Historically,yarn-dyed plaid fabrics(YDPFs)have enjoyed enduring popularity with many rich plaid patterns,but production data are still classified and searched only according to production parameters.The process does ...Historically,yarn-dyed plaid fabrics(YDPFs)have enjoyed enduring popularity with many rich plaid patterns,but production data are still classified and searched only according to production parameters.The process does not satisfy the visual needs of sample order production,fabric design,and stock management.This study produced an image dataset for YDPFs,collected from 10,661 fabric samples.The authors believe that the dataset will have significant utility in further research into YDPFs.Convolutional neural networks,such as VGG,ResNet,and DenseNet,with different hyperparameter groups,seemed themost promising tools for the study.This paper reports on the authors’exhaustive evaluation of the YDPF dataset.With an overall accuracy of 88.78%,CNNs proved to be effective in YDPF image classification.This was true even for the low accuracy of Windowpane fabrics,which often mistakenly includes the Prince ofWales pattern.Image classification of traditional patterns is also improved by utilizing the strip pooling model to extract local detail features and horizontal and vertical directions.The strip pooling model characterizes the horizontal and vertical crisscross patterns of YDPFs with considerable success.The proposed method using the strip pooling model(SPM)improves the classification performance on the YDPF dataset by 2.64%for ResNet18,by 3.66%for VGG16,and by 3.54%for DenseNet121.The results reveal that the SPM significantly improves YDPF classification accuracy and reduces the error rate of Windowpane patterns as well.展开更多
Body-fitted coordinate transformation equation was deduced and used to generate the body-fitted grids of molten pool for twin-roll strip casting.The orthogonality of the grids on the boundary was modified by adjusting...Body-fitted coordinate transformation equation was deduced and used to generate the body-fitted grids of molten pool for twin-roll strip casting.The orthogonality of the grids on the boundary was modified by adjusting source item.The energy equation and the boundary conditions were transformed from physical space to computational space.The velocity field model proposed by Hirohiko Takuda was used to calculate the temperature field of molten steel,and the influence of technical factors was also discussed.展开更多
A 1：1 water model of a twin-roll strip caster was set up based on the Froude number（Fr） and Reynolds number （Re） similarity criteria. Fluctuation and the mixing effect in the molten pool of twin-roll strip castin...A 1：1 water model of a twin-roll strip caster was set up based on the Froude number（Fr） and Reynolds number （Re） similarity criteria. Fluctuation and the mixing effect in the molten pool of twin-roll strip casting process were analyzed through hydraulic simulation. The height of the fluctuation of the liquid pool surface and the residence time were measured by DJ800 multifunctional monitor under different process conditions. The results show that the casting velocity, the insertion depth and the pool level have a great effect on the fluctuation.展开更多
随着遥感技术的发展,遥感图像的语义分割在城乡资源管理、城乡规划等领域有着更为广泛的应用。因为小型无人机在遥感数据采集方面具有成本效益、灵活性和操作便捷等优势,所以使用无人机拍摄图像已经成为收集遥感图像数据集的首选方法。...随着遥感技术的发展,遥感图像的语义分割在城乡资源管理、城乡规划等领域有着更为广泛的应用。因为小型无人机在遥感数据采集方面具有成本效益、灵活性和操作便捷等优势,所以使用无人机拍摄图像已经成为收集遥感图像数据集的首选方法。由于小型无人机低空斜角拍摄的特性,相较于传统遥感拍摄设备获取的图片,无人机图片目标细节信息更加丰富、目标关系更加复杂的特性导致基于局部卷积的传统深度学习模型无法再胜任此项工作。针对上述问题,提出了基于SegFormer的改进遥感图像语义分割网络。基于SegFormer,在编码层额外添加轮廓提取模块(edge contour extraction module,ECEM)辅助模型提取目标的浅层特征。鉴于城市遥感图像建筑物居多的特点,在编码层额外添加使用多尺度条纹池化(multi-scale strip pooling,MSP)替换全局平均池化的多尺度空洞空间卷积池化金字塔(multi-scale atrous spatial pyramid pooling,MSASPP)模块来提取图像中的长条状目标特征。针对原始解码器操作不利于特征信息还原的缺点,参考U-Net网络解码层的结构,将编码层接收到的特征融合之后再执行上采样提取以及SE通道注意力操作,以此加强特征的传播和融合。改进网络在国际摄影测量与遥感学会(International Society for Photogrammetry and Remote Sensing,ISPRS)提供的Vaihingen和无人机遥感图像语义分割数据集UAVid上进行了实验,网络分别取得了90.30%和77.90%的平均交并比(mean intersection over union,MIoU),比DeepLabV3+、Swin-Unet等通用分割网络具有更高的分割精确度。展开更多
近年来,随着深度学习的发展,在自然街景下的文本检测取得了巨大的进步,但在多方向和弯曲文本及对比度低的文本检测中的效果仍不理想。因此,针对弯曲文本和对比度低的文本的检测问题,提出了一种融合多尺度模块的文本检测方法,并通过检测...近年来,随着深度学习的发展,在自然街景下的文本检测取得了巨大的进步,但在多方向和弯曲文本及对比度低的文本检测中的效果仍不理想。因此,针对弯曲文本和对比度低的文本的检测问题,提出了一种融合多尺度模块的文本检测方法,并通过检测效果的提升,提高端到端文本识别的识别效果。针对RFB(Receptive Field Block)模块在下采样后局部信息丢失的问题,在RFB模块中嵌入极化自注意力(Polarized Self-Attention)机制以改进RFB来提取有效文本特征,提高特征图表征效果。针对特征金字塔(FPN)提取的特征不足、感受野小的问题,将改进的RFB模块嵌入特征金字塔(FPN)模块以增强特征提取融合。针对特征分布不确定性及远距离特征融合效果不佳的问题,引入条形池化(Strip Pooling)模块,进而提升检测方法的鲁棒性。在公开数据集Total-Text上的实验结果表明,该算法的F-measure值在端到端文本识别没有词汇表的情形下与目前高效的MaskTextSpotterV3相比高了0.3百分点,而在有词汇表的情形下则高出了0.2百分点;而在仅文本检测的情形下,该方法也有较为良好的表现。展开更多
基金This work was supported by China Social Science Foundation under Grant[17CG209]The fabric samples were supported by Jiangsu Sunshine Group and Jiangsu Lianfa Textile Group.
文摘Historically,yarn-dyed plaid fabrics(YDPFs)have enjoyed enduring popularity with many rich plaid patterns,but production data are still classified and searched only according to production parameters.The process does not satisfy the visual needs of sample order production,fabric design,and stock management.This study produced an image dataset for YDPFs,collected from 10,661 fabric samples.The authors believe that the dataset will have significant utility in further research into YDPFs.Convolutional neural networks,such as VGG,ResNet,and DenseNet,with different hyperparameter groups,seemed themost promising tools for the study.This paper reports on the authors’exhaustive evaluation of the YDPF dataset.With an overall accuracy of 88.78%,CNNs proved to be effective in YDPF image classification.This was true even for the low accuracy of Windowpane fabrics,which often mistakenly includes the Prince ofWales pattern.Image classification of traditional patterns is also improved by utilizing the strip pooling model to extract local detail features and horizontal and vertical directions.The strip pooling model characterizes the horizontal and vertical crisscross patterns of YDPFs with considerable success.The proposed method using the strip pooling model(SPM)improves the classification performance on the YDPF dataset by 2.64%for ResNet18,by 3.66%for VGG16,and by 3.54%for DenseNet121.The results reveal that the SPM significantly improves YDPF classification accuracy and reduces the error rate of Windowpane patterns as well.
文摘Body-fitted coordinate transformation equation was deduced and used to generate the body-fitted grids of molten pool for twin-roll strip casting.The orthogonality of the grids on the boundary was modified by adjusting source item.The energy equation and the boundary conditions were transformed from physical space to computational space.The velocity field model proposed by Hirohiko Takuda was used to calculate the temperature field of molten steel,and the influence of technical factors was also discussed.
文摘A 1：1 water model of a twin-roll strip caster was set up based on the Froude number（Fr） and Reynolds number （Re） similarity criteria. Fluctuation and the mixing effect in the molten pool of twin-roll strip casting process were analyzed through hydraulic simulation. The height of the fluctuation of the liquid pool surface and the residence time were measured by DJ800 multifunctional monitor under different process conditions. The results show that the casting velocity, the insertion depth and the pool level have a great effect on the fluctuation.
文摘随着遥感技术的发展,遥感图像的语义分割在城乡资源管理、城乡规划等领域有着更为广泛的应用。因为小型无人机在遥感数据采集方面具有成本效益、灵活性和操作便捷等优势,所以使用无人机拍摄图像已经成为收集遥感图像数据集的首选方法。由于小型无人机低空斜角拍摄的特性,相较于传统遥感拍摄设备获取的图片,无人机图片目标细节信息更加丰富、目标关系更加复杂的特性导致基于局部卷积的传统深度学习模型无法再胜任此项工作。针对上述问题,提出了基于SegFormer的改进遥感图像语义分割网络。基于SegFormer,在编码层额外添加轮廓提取模块(edge contour extraction module,ECEM)辅助模型提取目标的浅层特征。鉴于城市遥感图像建筑物居多的特点,在编码层额外添加使用多尺度条纹池化(multi-scale strip pooling,MSP)替换全局平均池化的多尺度空洞空间卷积池化金字塔(multi-scale atrous spatial pyramid pooling,MSASPP)模块来提取图像中的长条状目标特征。针对原始解码器操作不利于特征信息还原的缺点,参考U-Net网络解码层的结构,将编码层接收到的特征融合之后再执行上采样提取以及SE通道注意力操作,以此加强特征的传播和融合。改进网络在国际摄影测量与遥感学会(International Society for Photogrammetry and Remote Sensing,ISPRS)提供的Vaihingen和无人机遥感图像语义分割数据集UAVid上进行了实验,网络分别取得了90.30%和77.90%的平均交并比(mean intersection over union,MIoU),比DeepLabV3+、Swin-Unet等通用分割网络具有更高的分割精确度。
文摘近年来,随着深度学习的发展,在自然街景下的文本检测取得了巨大的进步,但在多方向和弯曲文本及对比度低的文本检测中的效果仍不理想。因此,针对弯曲文本和对比度低的文本的检测问题,提出了一种融合多尺度模块的文本检测方法,并通过检测效果的提升,提高端到端文本识别的识别效果。针对RFB(Receptive Field Block)模块在下采样后局部信息丢失的问题,在RFB模块中嵌入极化自注意力(Polarized Self-Attention)机制以改进RFB来提取有效文本特征,提高特征图表征效果。针对特征金字塔(FPN)提取的特征不足、感受野小的问题,将改进的RFB模块嵌入特征金字塔(FPN)模块以增强特征提取融合。针对特征分布不确定性及远距离特征融合效果不佳的问题,引入条形池化(Strip Pooling)模块,进而提升检测方法的鲁棒性。在公开数据集Total-Text上的实验结果表明,该算法的F-measure值在端到端文本识别没有词汇表的情形下与目前高效的MaskTextSpotterV3相比高了0.3百分点,而在有词汇表的情形下则高出了0.2百分点;而在仅文本检测的情形下,该方法也有较为良好的表现。
