Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inc...Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inconsistent from one view to another.This study develops a deep global multiple-scale and local patches attention(GMS-LPA)dual-branch network for pose-invariant FER to weaken the influence of pose variation and selfocclusion on recognition accuracy.In this research,the designed GMS-LPA network contains four main parts,i.e.,the feature extraction module,the global multiple-scale(GMS)module,the local patches attention(LPA)module,and the model-level fusion model.The feature extraction module is designed to extract and normalize texture information to the same size.The GMS model can extract deep global features with different receptive fields,releasing the sensitivity of deeper convolution layers to pose-variant and self-occlusion.The LPA module is built to force the network to focus on local salient features,which can lower the effect of pose variation and self-occlusion on recognition results.Subsequently,the extracted features are fused with a model-level strategy to improve recognition accuracy.Extensive experimentswere conducted on four public databases,and the recognition results demonstrated the feasibility and validity of the proposed methods.展开更多
Land consolidation (LC), as a type of human disturbance, improves land production efficiency and changes landscape distribution through land parcel reallocation. The objective of this study was to comparatively analyz...Land consolidation (LC), as a type of human disturbance, improves land production efficiency and changes landscape distribution through land parcel reallocation. The objective of this study was to comparatively analyze the changes of landscape patches before and after a land consolidation project (LCP) and the effects of land levelling, irrigation and drainage work and road engineering on the landscape structure. FRAGSTAT3.3 and buffer zone analysis were used to investigate those changes. The results suggest that the heterogeneity of landscape depressed, and tended to simplification after LC. Dry land was the most highly variable land use pattern, and the change of forestland was least due to its locations at a gradient larger than 25°. LC resulted in a more rational use of land, and could be an important step in promoting rural development in depressed and fragmented agricultural areas through unused land exploitation, small-patch combination, irrigation and water conservancy, and road construction. Land levelling leveled off the gradient field surface and decreased the slope. The fragmentized patches were much more incorporated with increasing slope. On the other hand, the ridge of a field became longer so that the length of field surface and area of patch were increased. Land levelling regulated, simplified and combined patches, so that the complexity degree was reduced. It is found that the buffer distance of 35 m was a turning point of human disturbance by irrigation and drainage systems, and patches presented flaky distribution when the buffer distance was smaller than 35 m. Meanwhile, the distance range between 25 m to 50 m was an impressible area for road engineering, which was sensitive to human actions, and the changes of all landscape metrics were larger than those in other buffer zones. In general, LC not only reallocated fragmented parcels, but also improved agricultural conditions.展开更多
最近,非局部滤波方法已成为滤波领域的研究热点.本文深入研究了基于预选择的非局部滤波方法,指出了已有方法在提取图像片特征方面存在的不足,利用二维主成分分析(Two-dimensional principal component analysis,2DPCA)提出了一种有效的...最近,非局部滤波方法已成为滤波领域的研究热点.本文深入研究了基于预选择的非局部滤波方法,指出了已有方法在提取图像片特征方面存在的不足,利用二维主成分分析(Two-dimensional principal component analysis,2DPCA)提出了一种有效的非局部滤波方法.该方法对基于预选择的非局部滤波方法的主要贡献有:1)用于提取图像片特征的面向图像片的2DPCA;2)基于相似距离直方图的相似集自动选取方法;3)相似距离权重参数局部自适应选取方法.实验结果表明,本文方法对弱梯度、人脸、纹理以及分段光滑图像均能取得较好的滤波效果.展开更多
基金supported by the National Natural Science Foundation of China (No.31872399)Advantage Discipline Construction Project (PAPD,No.6-2018)of Jiangsu University。
文摘Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inconsistent from one view to another.This study develops a deep global multiple-scale and local patches attention(GMS-LPA)dual-branch network for pose-invariant FER to weaken the influence of pose variation and selfocclusion on recognition accuracy.In this research,the designed GMS-LPA network contains four main parts,i.e.,the feature extraction module,the global multiple-scale(GMS)module,the local patches attention(LPA)module,and the model-level fusion model.The feature extraction module is designed to extract and normalize texture information to the same size.The GMS model can extract deep global features with different receptive fields,releasing the sensitivity of deeper convolution layers to pose-variant and self-occlusion.The LPA module is built to force the network to focus on local salient features,which can lower the effect of pose variation and self-occlusion on recognition results.Subsequently,the extracted features are fused with a model-level strategy to improve recognition accuracy.Extensive experimentswere conducted on four public databases,and the recognition results demonstrated the feasibility and validity of the proposed methods.
基金Funded by the Science and Technology Supporting Plan of China (No. 2006BAD05801-02)
文摘Land consolidation (LC), as a type of human disturbance, improves land production efficiency and changes landscape distribution through land parcel reallocation. The objective of this study was to comparatively analyze the changes of landscape patches before and after a land consolidation project (LCP) and the effects of land levelling, irrigation and drainage work and road engineering on the landscape structure. FRAGSTAT3.3 and buffer zone analysis were used to investigate those changes. The results suggest that the heterogeneity of landscape depressed, and tended to simplification after LC. Dry land was the most highly variable land use pattern, and the change of forestland was least due to its locations at a gradient larger than 25°. LC resulted in a more rational use of land, and could be an important step in promoting rural development in depressed and fragmented agricultural areas through unused land exploitation, small-patch combination, irrigation and water conservancy, and road construction. Land levelling leveled off the gradient field surface and decreased the slope. The fragmentized patches were much more incorporated with increasing slope. On the other hand, the ridge of a field became longer so that the length of field surface and area of patch were increased. Land levelling regulated, simplified and combined patches, so that the complexity degree was reduced. It is found that the buffer distance of 35 m was a turning point of human disturbance by irrigation and drainage systems, and patches presented flaky distribution when the buffer distance was smaller than 35 m. Meanwhile, the distance range between 25 m to 50 m was an impressible area for road engineering, which was sensitive to human actions, and the changes of all landscape metrics were larger than those in other buffer zones. In general, LC not only reallocated fragmented parcels, but also improved agricultural conditions.
文摘最近,非局部滤波方法已成为滤波领域的研究热点.本文深入研究了基于预选择的非局部滤波方法,指出了已有方法在提取图像片特征方面存在的不足,利用二维主成分分析(Two-dimensional principal component analysis,2DPCA)提出了一种有效的非局部滤波方法.该方法对基于预选择的非局部滤波方法的主要贡献有:1)用于提取图像片特征的面向图像片的2DPCA;2)基于相似距离直方图的相似集自动选取方法;3)相似距离权重参数局部自适应选取方法.实验结果表明,本文方法对弱梯度、人脸、纹理以及分段光滑图像均能取得较好的滤波效果.