Water bodies in urban areas are important as recreational areas. Thus, management plans that maintain high water quality are quite important. At the Hatadate Water Park adjacent to Miyagi University, water quality par...Water bodies in urban areas are important as recreational areas. Thus, management plans that maintain high water quality are quite important. At the Hatadate Water Park adjacent to Miyagi University, water quality parameters such as visibility, COD, TOC, and TN were monitored at a small pond and the inflowing stream from August to December in 2011, and photographs were taken of these sites. Variations in COD and TOC were highly related to changes in the physical appearance, especially changes in vegetation. These findings suggest: 1) the importance of management of vegetation for water quality control;and 2) the importance of collecting photographic records of sites for research purposes of interpreting data and even as a data point of water quality. Together with the water quality goals for water bodies in urban areas proposed by Sudo et al. [1], these water quality criteria were assessed, and it was notable that COD often exceeded the set goal. These results suggest that the maintenance of vegetation is more important than controlling incoming TN for primary production in the pond. Seasonal variations in COD and TOC were plotted for surface water of Kamafusa and Okura dams, both are important lakes in Miyagi area and the catchments of both lakes are mainly hilly area, using published water quality reports. Similar annual-cycle changing patterns were shown both for the dams, implying that some kinds of ecological factors in the catchments are affecting the water qualities of the dam, even at those larger scale water bodies. Finally, by shifting the focus from only water to upstream features such as small park, or pocket park, with a parking lot for the water body, the importance of landscape including vegetation and tree cover was highlighted.展开更多
Parsing of human images is a fundamental task for determining semantic parts such as the face,arms, and legs, as well as a hat or a dress. Recent deep-learning-based methods have achieved significant improvements, but...Parsing of human images is a fundamental task for determining semantic parts such as the face,arms, and legs, as well as a hat or a dress. Recent deep-learning-based methods have achieved significant improvements, but collecting training datasets with pixel-wise annotations is labor-intensive. In this paper,we propose two solutions to cope with limited datasets.Firstly, to handle various poses, we incorporate a pose estimation network into an end-to-end humanimage parsing network, in order to transfer common features across the domains. The pose estimation network can be trained using rich datasets and can feed valuable features to the human-image parsing network. Secondly, to handle complicated backgrounds,we increase the variation in image backgrounds automatically by replacing the original backgrounds of human images with others obtained from large-scale scenery image datasets. Individually, each solution is versatile and beneficial to human-image parsing, while their combination yields further improvement. We demonstrate the effectiveness of our approach through comparisons and various applications such as garment recoloring, garment texture transfer, and visualization for fashion analysis.展开更多
文摘Water bodies in urban areas are important as recreational areas. Thus, management plans that maintain high water quality are quite important. At the Hatadate Water Park adjacent to Miyagi University, water quality parameters such as visibility, COD, TOC, and TN were monitored at a small pond and the inflowing stream from August to December in 2011, and photographs were taken of these sites. Variations in COD and TOC were highly related to changes in the physical appearance, especially changes in vegetation. These findings suggest: 1) the importance of management of vegetation for water quality control;and 2) the importance of collecting photographic records of sites for research purposes of interpreting data and even as a data point of water quality. Together with the water quality goals for water bodies in urban areas proposed by Sudo et al. [1], these water quality criteria were assessed, and it was notable that COD often exceeded the set goal. These results suggest that the maintenance of vegetation is more important than controlling incoming TN for primary production in the pond. Seasonal variations in COD and TOC were plotted for surface water of Kamafusa and Okura dams, both are important lakes in Miyagi area and the catchments of both lakes are mainly hilly area, using published water quality reports. Similar annual-cycle changing patterns were shown both for the dams, implying that some kinds of ecological factors in the catchments are affecting the water qualities of the dam, even at those larger scale water bodies. Finally, by shifting the focus from only water to upstream features such as small park, or pocket park, with a parking lot for the water body, the importance of landscape including vegetation and tree cover was highlighted.
文摘Parsing of human images is a fundamental task for determining semantic parts such as the face,arms, and legs, as well as a hat or a dress. Recent deep-learning-based methods have achieved significant improvements, but collecting training datasets with pixel-wise annotations is labor-intensive. In this paper,we propose two solutions to cope with limited datasets.Firstly, to handle various poses, we incorporate a pose estimation network into an end-to-end humanimage parsing network, in order to transfer common features across the domains. The pose estimation network can be trained using rich datasets and can feed valuable features to the human-image parsing network. Secondly, to handle complicated backgrounds,we increase the variation in image backgrounds automatically by replacing the original backgrounds of human images with others obtained from large-scale scenery image datasets. Individually, each solution is versatile and beneficial to human-image parsing, while their combination yields further improvement. We demonstrate the effectiveness of our approach through comparisons and various applications such as garment recoloring, garment texture transfer, and visualization for fashion analysis.
