基于SF-Unet的高分辨率耕地遥感影像分割

利用深度学习实现遥感影像耕地区域自动化检测,取代人工解译,能有效提升耕地面积统计效率。针对目前存在分割目标尺度大且连续导致分割区域存在欠分割现象,边界区域情况复杂导致边缘分割困难等问题,提出了语义分割算法——Swin Transformer,TransFuse and U-Net(SF-Unet)。为强化网络不同层次特征提取和信息融合能力,提升边缘分割性能,使用U-Net网络替代TransFuse网络中的ResNet50模块;将Vision Transformer(ViT)替换为改进后的Swin Transformer网络,解决大区域的欠分割问题;通过注意力机制构建的Fusion融合模块将2个网络输出特征进行融合,增强模型对目标的语义表示,提高分割的精度。实验表明,SF-Unet语义分割网络在Gaofen Image Dataset(GID)数据集上的交并比(Intersection over Union,IoU)达到了90.57%,分别比U-Net和TransFuse网络提升了6.48%和6.09%,明显提升了耕地遥感影像分割的准确性。

2000-2017年江汉平原耕地遥感植被指数数据集

江汉平原是我国重要的农产区。本文利用MODIS卫星遥感数据,处理获得NDVI、EVI、DVI、RVI、SAVI、NDWI 6种遥感植被指数,整理为一个数据集。本数据集覆盖2000–2017年江汉平原的所有耕地区域,可作为江汉平原作物种植和生长情况的历史溯源研究和江汉平原农田生态演进研究的重要基础数据。

基于遥感影像耕地解译样本库的构建及解译方法研究

现如今,遥感技术正在飞速的发展,国内生产的遥感卫星技术辨识率也逐渐上升,提取数据的频率与周期也在稳步的上升,针对遥感影像信息的提取和变化监测提出新型要求。故此,本文从遥感影像耕地解译样本库的构建及解译方法分析,并采用相关系统,以此助力于探寻遥感影像耕地解译样本库提取与变化监测实践应用的新型高效方法。

Dynamic of Chinas cultivated land and landcover changes of its typical regions based on remote sensing data

Using the multi-temporal Landsat data and survey data of national resources, the authors studied the dynamics of cultivated land and landcover changes of typical ecological regions in China. The results of investigation showed that the whole distribution of the cultivated land shifted to Northeast and Northwest China, and as a result, the ecological quality of cultivated land dropped down. The seacoast and cultivated land in the area of Yellow River Mouth expanded by an increasing rate of 0.73 kma-1, with a depositing rate of 2.1 kma-1. The desertification area of the dynamic of Horqin Sandy Land increased from 60.02% of the total land area in1970s to 64.82% in1980s but decreased to 54.90% in early 1990s. As to the change of North Tibet lakes, the water area of the Namu Lake decreased by 38.58 km2 from year 1970 to 1988, with a decreasing rate of 2.14 km2a-1.

Cultivated Land Changes and Their Driving Forces——A Satellite Remote Sensing Analysis in the Yellow River Delta,China

Taking Kenli County in the Yellow River Delta, China, as the study area and using digital satellite remote sensing techniques, cultivated land use changes and their corresponding driving forces were explored in this study. An interactive interpretation and a manual modification procedure were carried out to acquire cultivated land information. An overlay method based on classification results and a visual change detection method which was supported by land use maps were employed to detect the cultivated land changes. Based on the changes that were revealed and a spatial analysis between cultivated land use and related natural and socio-economic factors, the driving forces for cultivated land use changes in the study area were determined.The results showed a decrease in cultivated land in Kenli County of 5321.8 ha from 1987 to 1998, i.e.,an average annual decrement of 483.8 ha, which occurred mainly in the central paddy field region and the northeast dry land region. Adverse human activities, soil salinization and water deficiencies were the driving forces that caused these cultivated land use changes.

Cropland Dynamics and Their Influence on the Productivity in Northern Shaanxi,China,for the Past 20 Years:Based on Remotely Sensed Data

In this paper,we mainly focused our research on northern Shaanxi district,which is a pilot area of the Grain for Green Project.We compared the spatial distribution patterns of croplands and their productivity for the past 20 years（from the end of the 1980 s to 2010）.Cropland dynamics for the past 20 years were interpreted from medium- and high-resolution remote sensing images（Landsat TM/ETM＋）.In addition,using the GLO-PEM and AGRO-VPM models with a medium resolution and long time series remote sensing dataset（AVHRR/MODIS）,net primary productivity（NPP） and its relationship with cropland were estimated.Finally,the effect of cropland change on productivity was analyzed.The results show that during the first decade of the research period,cropland area and productivity in northern Shaanxi experienced a small boost,while in the latter decade,both cropland area and NPP were significantly reduced.The main cause of the increase in cropland was the reclamation of large area of grassland and unutilized land to meet the food demands of the local population as well as to compensate for the occupation of urban constructions.While the main cause of the decrease in cropland was the implementation of the Grain for Green Project.In addition,urbanization was also a key factor.Overall,during the past 20 years,the total area of cropland in northern Shaanxi decreased by 42.56%,and cropland NPP dropped by 41.90%.This study is of great importance for the assessment of regional cropland security,food security and scientific planning of regional land use.

Patterns and Driving Forces of Cropland Abandonment in Mountainous Areas

Cropland abandonment is spreading from developed countries to developing countries such as China.Cropland abandonment in China commonly occurs in mountainous areas due to their specific natural and geographical conditions.However,due to the lack of dependable monitoring methods via medium-high-resolution remote sensing images,the scale of abandoned cropland in many mountainous areas of China is unclear,and the mechanisms driving cropland abandonment have not been clearly identified.To overcome these limitations,we took Zhong County of Chongqing in China as an example,and used Landsat 8 OLI_TIRS remote sensing image data to develop a method for mapping abandoned cropland in mountainous areas based on annual land use change monitoring.At the same time,the ridge regression method was adopted to analyze the factors influencing cropland abandonment.These analyses showed that the cropland abandonment rate in Zhong County of Chongqing was as high as 7.86%,while the overall accuracy of identifying abandoned cropland was as high as 90.82%.Among the social and economic factors that affect cropland abandonment,the rural population,economic development,and livestock husbandry development were the most important ones.At the land parcel scale,large-scale cropland abandonment occurred in areas at elevations above 650 m or with slopes of more than 15°.

Chinese Cropland Quality and Its Temporal and Spatial Changes due to Urbanization in 2000–2015

The acceleration of urbanization has led to the occupation of more cropland, especially higher quality cropland, which could pose a huge threat to food security and have other implications for the inadequate cropland resource supply in China. Though the spatial status of Chinese cropland quality has been assessed, its temporal changes since 2000 to 2015 are still not clear. An accumulated probability distribution method was used to determine the criteria of cropland quality using the net primary production data product(MOD17) from Moderate Resolution Imaging Spectroradiometer(MODIS). Then the cropland quality of higher, median and lower production was spatially mapped and its changes due to occupation by urbanization were analyzed through the land use changes(LUCC) data primarily from Landsat TM images in the three periods of 2000–2005, 2005–2010, and 2010–2015.The results showed that of the total cropland reduction area the proportion taken by urbanization increased from47.29% in the early stage to 77.46% in the recent period. The quality of Chinese cropland was dominated by lowand medium-yield fields, accounting for 40.81% and 48.74%, respectively, with high-yield fields accounting for only 10.44% of the total cropland in the country in 2000. The high-yield areas have been seriously threatened by the expansion of construction land fields, with the ratio of high-yield area to total area occupied by urbanization increasing from 9.71% in 2000–2005 to 15.63% in 2010–2015. Spatially, this phenomenon has been moving from eastern and southern China to central and western China, especially in Northwest China where the ratio has arrived at the highest proportion, with 52.97% of high-yield cropland in the total land taken by the expansion by 2015. This study not only provides a method to assess cropland quality but also reveals the threatening trend from the expansion of urbanization on high-quality cropland. More attention should be paid to the latter in land use planning and policies made to prevent threats to food security from declines in both cropland quantity and quality.