Soil seed bank refers to the sum of upper litter decay and all survival seeds in the soil. It plays a unique role in restoring the surface vegetation and protecting biodiversity. Chinese scholars have carried out exte...Soil seed bank refers to the sum of upper litter decay and all survival seeds in the soil. It plays a unique role in restoring the surface vegetation and protecting biodiversity. Chinese scholars have carried out extensive researches on the soil seed bank,and have obtained considerable achievements. On the basis of reading extensive literature,this paper summarized the current research results and found that the research contents of soil seed bank in China mainly involve the following aspects:( i) research method of soil seed bank;( ii) classification system of soil seed bank;( iii) spatial pattern of soil seed bank;( iv) relationship between soil seed bank and surface vegetation;( v) dynamic characteristic and seedling replanting of soil seed bank;( vi) application of soil seed bank in vegetation restoration. On the basis of the above summary,it came up with existing problems and future research direction of China's soil seed bank from system construction,field observation,disciplinary combination,interference factors,and community succession.展开更多
The aim of this paper is to investigate the characteristics of seed banks in six land use types of urban wetland in Hangzhou Xixi. The results show that the seed bank density of abandoned farmland and farmland are mor...The aim of this paper is to investigate the characteristics of seed banks in six land use types of urban wetland in Hangzhou Xixi. The results show that the seed bank density of abandoned farmland and farmland are more than the others. The seed banks density of the uprightness in abandoned farmland, pond bank I, II and bamboo non-timber forest are downtrend, which is different from the pattern of low-high-low in farmland, and there is no difference among the density of three lays in man-made landscape area. There are 35 families and 102 species totally according to six different land use types of urban wetland in Hangzhou Xixi. A part of its seed bank types obtain the seeds of shrub, which the annual and perennial seeds use the interval of early spring to germinate as the same as farmland type. Farmland type also owns another small part of the seed banks which germinate in Autumn. The seed banks type of bamboo non-timber forest and pond bank are the herbage and shrub type germinating under the condition of drought and interference with the contribution of permanent seed banks.展开更多
目的:提出一种基于切片滤波器组共空间模式(slice filter bank common spatial pattern,SFBCSP)的运动想象分类方法,降低脑电运动想象数据集之间的系统性差异对分类模型的影响,提升分类准确率。方法:基于SFBCSP进行运动想象分类,首先对...目的:提出一种基于切片滤波器组共空间模式(slice filter bank common spatial pattern,SFBCSP)的运动想象分类方法,降低脑电运动想象数据集之间的系统性差异对分类模型的影响,提升分类准确率。方法:基于SFBCSP进行运动想象分类,首先对原始脑电信号进行预处理,其次通过带通滤波器和滑动窗口将脑电信号切分为不同频带和时间片段,然后利用空间滤波器提取脑电信号不同电极通道之间的空域特征,最后对多维度特征进行筛选和分类。利用神经信息处理系统大会提供的竞赛数据集对SFBCSP方法与脑电神经网络EEGNet、卷积循环注意力模型(convolutional recurrent attention model,CRAM)、共空间模式(common spatial pattern,CSP)和滤波器组共空间模式(filter bank common spatial pattern,FBCSP)的分类准确率进行比较。结果:提出的SFBCSP方法平均分类准确率均高于EEGNet、CRAM、CSP和FBCSP,相较于FBCSP,SFBCSP在单个数据集、跨数据集和混合数据集上的平均分类准确率分别提升了7.26%、5.16%和1.96%。结论:所提出的SFBCSP方法能够降低跨数据集的系统性差异对分类模型的影响,对运动想象和脑机接口的研究具有重要意义。展开更多
基金Supported by Project of Natural Science Foundation of Tibet Autonomous Region(2016ZR-15-40)
文摘Soil seed bank refers to the sum of upper litter decay and all survival seeds in the soil. It plays a unique role in restoring the surface vegetation and protecting biodiversity. Chinese scholars have carried out extensive researches on the soil seed bank,and have obtained considerable achievements. On the basis of reading extensive literature,this paper summarized the current research results and found that the research contents of soil seed bank in China mainly involve the following aspects:( i) research method of soil seed bank;( ii) classification system of soil seed bank;( iii) spatial pattern of soil seed bank;( iv) relationship between soil seed bank and surface vegetation;( v) dynamic characteristic and seedling replanting of soil seed bank;( vi) application of soil seed bank in vegetation restoration. On the basis of the above summary,it came up with existing problems and future research direction of China's soil seed bank from system construction,field observation,disciplinary combination,interference factors,and community succession.
文摘The aim of this paper is to investigate the characteristics of seed banks in six land use types of urban wetland in Hangzhou Xixi. The results show that the seed bank density of abandoned farmland and farmland are more than the others. The seed banks density of the uprightness in abandoned farmland, pond bank I, II and bamboo non-timber forest are downtrend, which is different from the pattern of low-high-low in farmland, and there is no difference among the density of three lays in man-made landscape area. There are 35 families and 102 species totally according to six different land use types of urban wetland in Hangzhou Xixi. A part of its seed bank types obtain the seeds of shrub, which the annual and perennial seeds use the interval of early spring to germinate as the same as farmland type. Farmland type also owns another small part of the seed banks which germinate in Autumn. The seed banks type of bamboo non-timber forest and pond bank are the herbage and shrub type germinating under the condition of drought and interference with the contribution of permanent seed banks.
基金国家自然科学基金项目(30700094)Research project for a sustainable development of economic and social structure dependent on the environment of the eastern coast of Asia from Tokyo University of Information
文摘目的:提出一种基于切片滤波器组共空间模式(slice filter bank common spatial pattern,SFBCSP)的运动想象分类方法,降低脑电运动想象数据集之间的系统性差异对分类模型的影响,提升分类准确率。方法:基于SFBCSP进行运动想象分类,首先对原始脑电信号进行预处理,其次通过带通滤波器和滑动窗口将脑电信号切分为不同频带和时间片段,然后利用空间滤波器提取脑电信号不同电极通道之间的空域特征,最后对多维度特征进行筛选和分类。利用神经信息处理系统大会提供的竞赛数据集对SFBCSP方法与脑电神经网络EEGNet、卷积循环注意力模型(convolutional recurrent attention model,CRAM)、共空间模式(common spatial pattern,CSP)和滤波器组共空间模式(filter bank common spatial pattern,FBCSP)的分类准确率进行比较。结果:提出的SFBCSP方法平均分类准确率均高于EEGNet、CRAM、CSP和FBCSP,相较于FBCSP,SFBCSP在单个数据集、跨数据集和混合数据集上的平均分类准确率分别提升了7.26%、5.16%和1.96%。结论:所提出的SFBCSP方法能够降低跨数据集的系统性差异对分类模型的影响,对运动想象和脑机接口的研究具有重要意义。
