关键词：
deep learning;convolutional neural network;tree species classification;random forest;OHS-1 hyperspectral image
摘要：
The accurate characterization of tree species distribution in forest areas can help significantly reduce uncertainties in the estimation of ecosystem parameters and forest resources. Deep learning algorithms have become a hot topic in recent years, but they have so far not been applied to tree species classification. In this study, one-dimensional convolutional neural network (Conv1D), a popular deep learning algorithm, was proposed to automatically identify tree species using OHS-1 hyperspectral images. Additionally, the random forest (RF) classifier was applied to compare to the algorithm of deep learning. Based on our experiments, we drew three main conclusions: First, the OHS-1 hyperspectral images used in this study have high spatial resolution (10 m), which reduces the influence of mixed pixel effect and greatly improves the classification accuracy. Second, limited by the amount of sample data, Conv1D-based classifier does not need too many layers to achieve high classification accuracy. In addition, the size of the convolution kernel has a great influence on the classification accuracy. Finally, the accuracy of Conv1D (85.04%) is higher than that of RF model (80.61%). Especially for broadleaf species with similar spectral characteristics, such as Manchurian walnut and aspen, the accuracy of Conv1D-based classifier is significantly higher than RF classifier (87.15% and 71.77%, respectively). Thus, the Conv1D-based deep learning framework combined with hyperspectral imagery can efficiently improve the accuracy of tree species classification and has great application prospects in the future.
作者机构：
[Wangzhengqing; Liushan; Huyang; Xieyanshi; Duanxianzhe; Fengzhigang; Chenliang] Univ South China, Sch Nucl Resources Engn, Hengyang 421001, Hunan, Peoples R China.;[Huyang] Univ South China, Postdoctoral Res Stn Min Engn, Hengyang 421001, Hunan, Peoples R China.;[Huyang] Cooperat Innovat Ctr Nucl Fuel Cycle Technol & Eq, Hengyang 421001, Hunan, Peoples R China.;[Jiazhikun] Sinochem Petr Explorat & Prod Co Ltd, Beijing 100031, Peoples R China.
会议名称：
4th International Conference on Energy and Environmental Protection (ICEEP)
会议时间：
JUN 02-04, 2015
会议地点：
Shenzhen, PEOPLES R CHINA
会议主办单位：
[Huyang;Fengzhigang;Xieyanshi;Chenliang;Duanxianzhe;Wangzhengqing;Liushan] Univ South China, Sch Nucl Resources Engn, Hengyang 421001, Hunan, Peoples R China.^[Huyang] Univ South China, Postdoctoral Res Stn Min Engn, Hengyang 421001, Hunan, Peoples R China.^[Huyang] Cooperat Innovat Ctr Nucl Fuel Cycle Technol & Eq, Hengyang 421001, Hunan, Peoples R China.^[Jiazhikun] Sinochem Petr Explorat & Prod Co Ltd, Beijing 100031, Peoples R China.
关键词：
sandstone-type uranium deposit;chemical speciation of uranium;sequential chemical extraction;in-situ Leaching
摘要：
This paper presents a method of studying uranium speciation with six different hole depth and numbers from in-situ leaching sandstone-type uranium deposits by a sequential extraction procedure and demonstrates its application to sandstone uranium exploration. The chemical extraction procedure was modified from Tessier. The chemical speciation of uranium in samples was classified into six speciation: exchangeable ions, bound to carbonates, bound to sulfide-organic matter, bound to amorphous Fe-Mn oxides/bound to hydroxide, bound to sparry Fe-Mn oxides/bound to hydroxide and residual speciation. Studying the chemical speciation of uranium by method of sequential extraction show that the uranium distribution characteristics were significantly different whether they were in different samples or in the same sample. Therefore, a research on chemical speciation of Deposit behind the production is essential to reasonable evaluation of Deposit and guide the technology of in-situ leaching uranium Deposit. It also can rich the mineralization mechanism of sandstone-type uranium Deposit. The average amounts of bound to carbonates, exchangeable ions, bound to amorphous Fe-Mn oxides, residual speciation, bound to sulfide-organic matter and bound to sparry Fe-Mn oxides in order as 58.9%, 22.6%, 13.3%, 2.3%, 1.4% and 1.4%.
