Based on the statistical characteristics of remote sensing data, the spatial geometric structure characteristics of spectral data and distribution of background, interference and alteration information in characterist...Based on the statistical characteristics of remote sensing data, the spatial geometric structure characteristics of spectral data and distribution of background, interference and alteration information in characteristic space were researched through the analysis of two-dimensional and three-dimensional scatter diagrams. The results indicate that the hyper-space of remote sensing multi-data aggregation belongs to low-dimensional geometric structure, i.e. hyperplane form, and anomalous point groups including alteration information usually dissociate out of hyperplane. Scatter diagrams of remote sensing data band are mainly presented as two distribution forms of single-ellipse and dual-ellipse. Clarifying the relations of three objects of background, disturbance and alteration information in remote sensing images provides an important technical thought and guidance for accurately detecting and extracting remote sensing alteration information.展开更多
Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area ...Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area of the temperate zone on the globe. The well reserved primeval vertical distribution of natural landscape belts and the Natural Conservation of Changbai Mountains adopted by the UNESCO′s MAB Program cause the worldwide attention of geographers. Beside the complexity of the climatic structure itself, the mechanical effection of the high mountain body also effect the climate in the eastern part of China. In the mountain area where short of meteorological observation data, the climatic study by remote sensing is favorable for discovery and representation of climatic law in large area.展开更多
The Soil Conservation Monitorins Information System (SCMIS) presented in this paper is oriented to soil erosion control, resources exploitation, utilization, planning and management for a small watershed (about 10 sq....The Soil Conservation Monitorins Information System (SCMIS) presented in this paper is oriented to soil erosion control, resources exploitation, utilization, planning and management for a small watershed (about 10 sq. km.) on the Loess Plateau. It sums up Remote sensing (RS), Geographical Information System (GIS) and Expert System (ES) and consists of a integrated system. As a basic level information system of Loess Plateau, its perfection and psreading will bring about a great advance in resources exploitation and management of Loess Plateau.展开更多
Accurate estimation of soil lead pollution degree is one of the key steps in controlling soil lead pollution; vegetable hyperspectral features research provided a new approach to discovering and monitoring soil heavy ...Accurate estimation of soil lead pollution degree is one of the key steps in controlling soil lead pollution; vegetable hyperspectral features research provided a new approach to discovering and monitoring soil heavy metal pollution.Spectral reflectance implies information of pollution impacts on vegetation;estimation of lead pollution degree based on the spectral reflectance is equivalent to extraction of weak information.This study puts forward a new feature extraction method based展开更多
针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积...针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积,增强算法对样本空间特征的捕捉能力。其次,采用Ghost Convolution结合跳跃连接和特征分支策略设计多层级信息增强模块,增强特征整合。随后引入级联注意力CGA(cascaded group attention),通过组内独立注意力计算,加强模型对多样化地物形态的适应性。最后,通过动态深度特征增强器构造特征融合模块,进一步加强模型捕获能力。在WHU数据集上实验结果表明:改进算法较基线模型F1-Score提高8.57%,mIoU提高12.48%,与其他主流语义分割模型相比,改进DeepLabv3+具有更好的分割精度。展开更多
针对遥感图像微小目标检测中存在的浅层细化特征、深层语义表征和多尺度信息提取3个问题,提出一种综合运用多项技术的跨尺度YOLOv7(cross-scale YOLOv7,CSYOLOv7)网络。首先,设计跨阶段特征提取模块(cross-stage feature extraction mod...针对遥感图像微小目标检测中存在的浅层细化特征、深层语义表征和多尺度信息提取3个问题,提出一种综合运用多项技术的跨尺度YOLOv7(cross-scale YOLOv7,CSYOLOv7)网络。首先,设计跨阶段特征提取模块(cross-stage feature extraction module,CFEM)和感受野特征增强模块(receptive field feature enhancement module,RFFEM)。CFEM提高模型细化特征提取能力并抑制浅层下采样过程中特征的丢失,RFFEM加大网络对深层语义特征的提取力度,增强模型对目标上下文信息获取能力。其次,设计跨梯度空间金字塔池化模块(cross-gradient space pyramid pool module,CSPPM)有效融合微小目标多尺度的全局和局部特征。最后,用形状感知交并比(shape-aware intersection over union,Shape IoU)替换完全交并比(complete intersection over union,CIoU),提高模型在边界框定位任务中的精确度。实验结果表明,CSYOLOv7网络在DIOR(dataset for image object recognition)数据集和NWPU VHR-10(Northwestern Polytechnical University Very High Resolution-10)数据集上分别取得了74%和89.6%的检测精度,有效提升遥感图像微小目标的检测效果。展开更多
基金Project(2006BAB01A06) supported by the National Science and Technology Pillar Program Project during the 11th Five-Year Plan PeriodProject(1212010761503) supported by Land and Resources Investigation Project
文摘Based on the statistical characteristics of remote sensing data, the spatial geometric structure characteristics of spectral data and distribution of background, interference and alteration information in characteristic space were researched through the analysis of two-dimensional and three-dimensional scatter diagrams. The results indicate that the hyper-space of remote sensing multi-data aggregation belongs to low-dimensional geometric structure, i.e. hyperplane form, and anomalous point groups including alteration information usually dissociate out of hyperplane. Scatter diagrams of remote sensing data band are mainly presented as two distribution forms of single-ellipse and dual-ellipse. Clarifying the relations of three objects of background, disturbance and alteration information in remote sensing images provides an important technical thought and guidance for accurately detecting and extracting remote sensing alteration information.
文摘Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area of the temperate zone on the globe. The well reserved primeval vertical distribution of natural landscape belts and the Natural Conservation of Changbai Mountains adopted by the UNESCO′s MAB Program cause the worldwide attention of geographers. Beside the complexity of the climatic structure itself, the mechanical effection of the high mountain body also effect the climate in the eastern part of China. In the mountain area where short of meteorological observation data, the climatic study by remote sensing is favorable for discovery and representation of climatic law in large area.
文摘The Soil Conservation Monitorins Information System (SCMIS) presented in this paper is oriented to soil erosion control, resources exploitation, utilization, planning and management for a small watershed (about 10 sq. km.) on the Loess Plateau. It sums up Remote sensing (RS), Geographical Information System (GIS) and Expert System (ES) and consists of a integrated system. As a basic level information system of Loess Plateau, its perfection and psreading will bring about a great advance in resources exploitation and management of Loess Plateau.
文摘Accurate estimation of soil lead pollution degree is one of the key steps in controlling soil lead pollution; vegetable hyperspectral features research provided a new approach to discovering and monitoring soil heavy metal pollution.Spectral reflectance implies information of pollution impacts on vegetation;estimation of lead pollution degree based on the spectral reflectance is equivalent to extraction of weak information.This study puts forward a new feature extraction method based
文摘针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积,增强算法对样本空间特征的捕捉能力。其次,采用Ghost Convolution结合跳跃连接和特征分支策略设计多层级信息增强模块,增强特征整合。随后引入级联注意力CGA(cascaded group attention),通过组内独立注意力计算,加强模型对多样化地物形态的适应性。最后,通过动态深度特征增强器构造特征融合模块,进一步加强模型捕获能力。在WHU数据集上实验结果表明:改进算法较基线模型F1-Score提高8.57%,mIoU提高12.48%,与其他主流语义分割模型相比,改进DeepLabv3+具有更好的分割精度。
文摘针对遥感图像微小目标检测中存在的浅层细化特征、深层语义表征和多尺度信息提取3个问题,提出一种综合运用多项技术的跨尺度YOLOv7(cross-scale YOLOv7,CSYOLOv7)网络。首先,设计跨阶段特征提取模块(cross-stage feature extraction module,CFEM)和感受野特征增强模块(receptive field feature enhancement module,RFFEM)。CFEM提高模型细化特征提取能力并抑制浅层下采样过程中特征的丢失,RFFEM加大网络对深层语义特征的提取力度,增强模型对目标上下文信息获取能力。其次,设计跨梯度空间金字塔池化模块(cross-gradient space pyramid pool module,CSPPM)有效融合微小目标多尺度的全局和局部特征。最后,用形状感知交并比(shape-aware intersection over union,Shape IoU)替换完全交并比(complete intersection over union,CIoU),提高模型在边界框定位任务中的精确度。实验结果表明,CSYOLOv7网络在DIOR(dataset for image object recognition)数据集和NWPU VHR-10(Northwestern Polytechnical University Very High Resolution-10)数据集上分别取得了74%和89.6%的检测精度,有效提升遥感图像微小目标的检测效果。
