Quantitative data are essential to an appro-priate characterization of vegetation.In the past few years,considerable attention has been paid to vegetation sampling techniques.A number of methods have been developed fo...Quantitative data are essential to an appro-priate characterization of vegetation.In the past few years,considerable attention has been paid to vegetation sampling techniques.A number of methods have been developed for plant density estimations that utilize spacing distances instead of fixed-area quadrats.In this paper,we review the main distance methods for estimating density and propose a new distance method denominated the quartered neighbor method.In this method,the sampling point is considered the center,and the area around it is divided into four quadrants.The distance from the closest individual in each quadrant to its closest neighbor in the same quadrant is measured,and the average of them is the distance we need.It is actually an integration of two old distance methods,the nearest neighbor method,and the point-centered quarter method.With our new method and an old distance method(the point-centered quarter method),we calculated the average spacing distances of the Larix principis-rupprechtii population in the larch forests of the Donglingshan Mountain.Comparing the two methods with the quadrat method,we found they were almost the same in accuracy,but the precision of the new one was better.Meanwhile,it is adequate in sampling intensity and adaptable for general use in rapid ecological survey work.展开更多
因传统方法单一使用遥感影像或兴趣点(point of interest,POI)数据识别城市功能区时,存在城市特征信息利用不完全、识别精度不高的问题,提出利用POI数据、遥感影像等多源数据,并将自然特征和人文特征相结合,采用基于掩膜区域卷积神经网...因传统方法单一使用遥感影像或兴趣点(point of interest,POI)数据识别城市功能区时,存在城市特征信息利用不完全、识别精度不高的问题,提出利用POI数据、遥感影像等多源数据,并将自然特征和人文特征相结合,采用基于掩膜区域卷积神经网络和样方密度法(mask region based convolutional neural network and quadrat density method,Mask R-CNN-QDM)模型识别城市功能区的方法。首先基于遥感影像采用Mask R-CNN模型识别建筑物,然后将识别结果与POI数据进行补充校验,得到结合自然特征和人文特征的分类结果,再引入面积要素对分类结果进行赋分,以计算样方密度,并采用随机抽样方式对所提方法功能区的识别精度进行评价。研究结果表明,Mask R-CNN-QDM模型的识别精确度高达0.900,平均Kappa系数为0.802,说明该方法能较好地区分单一城市功能区和混合城市功能区。展开更多
基金Project (51175138) supported by the National Natural Science Foundation of ChinaProject (20100111110003) supported by Specialized Research Fund for the Doctoral Program of Higher Education, ChinaProject (10040606Y21) supported by the Science and Technological Fund of Anhui Province for Outstanding Youth, China
基金The study was supported by the National Natural Science Foundation of China(Grant No.30870399)the Teachers’Foundation of Education Ministry of China.
文摘Quantitative data are essential to an appro-priate characterization of vegetation.In the past few years,considerable attention has been paid to vegetation sampling techniques.A number of methods have been developed for plant density estimations that utilize spacing distances instead of fixed-area quadrats.In this paper,we review the main distance methods for estimating density and propose a new distance method denominated the quartered neighbor method.In this method,the sampling point is considered the center,and the area around it is divided into four quadrants.The distance from the closest individual in each quadrant to its closest neighbor in the same quadrant is measured,and the average of them is the distance we need.It is actually an integration of two old distance methods,the nearest neighbor method,and the point-centered quarter method.With our new method and an old distance method(the point-centered quarter method),we calculated the average spacing distances of the Larix principis-rupprechtii population in the larch forests of the Donglingshan Mountain.Comparing the two methods with the quadrat method,we found they were almost the same in accuracy,but the precision of the new one was better.Meanwhile,it is adequate in sampling intensity and adaptable for general use in rapid ecological survey work.
文摘因传统方法单一使用遥感影像或兴趣点(point of interest,POI)数据识别城市功能区时,存在城市特征信息利用不完全、识别精度不高的问题,提出利用POI数据、遥感影像等多源数据,并将自然特征和人文特征相结合,采用基于掩膜区域卷积神经网络和样方密度法(mask region based convolutional neural network and quadrat density method,Mask R-CNN-QDM)模型识别城市功能区的方法。首先基于遥感影像采用Mask R-CNN模型识别建筑物,然后将识别结果与POI数据进行补充校验,得到结合自然特征和人文特征的分类结果,再引入面积要素对分类结果进行赋分,以计算样方密度,并采用随机抽样方式对所提方法功能区的识别精度进行评价。研究结果表明,Mask R-CNN-QDM模型的识别精确度高达0.900,平均Kappa系数为0.802,说明该方法能较好地区分单一城市功能区和混合城市功能区。
