Accurate urban land-use mapping is a challenging task in the remote-sensing field. With availability of diverse remote sensors, synthetic use and integration multisource data provides an opportunity for improving classification accuracy. Neural networks Deep Learning have achieved very promising results computer-vision tasks, such as image object detection. However, problem designing effective deep-learning model fusion still remains. To tackle this issue, paper proposes modified two-branch...

Coastal land cover classification is a significant yet challenging task in remote sensing because of the complex and fragmented nature coastal landscapes. However, availability multitemporal multisensor data provides opportunities to improve accuracy. Meanwhile, rapid development deep learning has achieved astonishing results computer vision tasks also been popular topic field sensing. Nevertheless, designing an effective concise model for remains problematic. To tackle this issue, we...

With the rapid urbanization process in China, numerous urban villages have been appeared, which are surrounded by newly-built blocks. Due to high population density, poor hygiene, chaotic waste discharge, and inadequate public facilities, many negative impacts on both environment management. The objective of this study is propose a dual-branch deep learning model for multi-modal satellite street-view data fusion detect Beijing, Tianjin Shijiazhuang, core cities Jing-Jin-Ji region China....

The spatial distribution of crops is an important agricultural parameter, which used to derive information about crop productivity and food security. However, mapping on a large scale challenging due the low spatio-temporal satellite data, sparse sampling, poor computational efficiency for massive data. To alleviate these problems, this study proposes method based discrete grids with machine learning integrate GaoFen-1 Sentinel-2 imagery. First, proposed fuses multi-source data similar...

Plantation is an important land use type that differs from natural forests and affects the economy environment. Tree age one of key factors used to quantify impact plantations. However, there a lack datasets explicitly documenting planting years global Here we time-series Landsat archive 1982 2020 LandTrendr algorithm generate maps based on plantation extent products in Google Earth Engine (GEE) platform. The developed this study are GeoTIFF format with 30-meter spatial resolution by...

Crop mapping using satellite imagery is crucial for agriculture applications. However, a fundamental challenge that hinders crop progress the scarcity of samples. The latest foundation model, Segment Anything Model (SAM), provides an opportunity to address this issue, yet few studies have been conducted in area. This study investigated parcel segmentation performance SAM on commonly used medium-resolution (i.e., Sentinel-2 and Landsat-8) proposed novel automated sample generation framework...

Vegetable mapping from remote sensing imagery is important for precision agricultural activities such as automated pesticide spraying. Multi-temporal unmanned aerial vehicle (UAV) data has the merits of both very high spatial resolution and useful phenological information, which shows great potential accurate vegetable classification, especially under complex fragmented landscapes. In this study, an attention-based recurrent convolutional neural network (ARCNN) been proposed multi-temporal...

The greenhouse is the fastest growing food production approach and has become symbol of protected agriculture with development agricultural modernization. Previous studies have verified effectiveness remote sensing techniques for mono-temporal mapping. In practice, long-term monitoring from data vital sustainable management existing been limited in understanding its spatiotemporal dynamics. This study aimed to generate multi-temporal maps a typical region (Shouguang region, north China) 1990...

As the important components of modern facility agriculture, both plastic greenhouses and mulching films have been widely utilized in agriculture production. Due to similarity spectral signatures, it remains a challenging task separate from each other. Meanwhile, deep learning has achieved great performance many computer vison tasks, become research hotspot remote sensing image analysis. However, rarely studied for accurate mapping agricultural covers, especially long-neglected issue...

Economic fruit forest is an important part of Chinese agriculture with high economic value and ecological benefits. Using UAV multi-spectral images to research the classification forests based on deep learning great significance for accurately understanding distribution scale status quo national resources. Based remote sensing UAV, this paper constructed semantic segmentation data forests, conducted a comparative study identification FCN, SegNet, U-Net classic models, proposed improved...

Remote sensing and land resource surveys have been used in recent decades for use/land cover (LULC) mapping; however, keeping the developed LULC up-to-date consistent with survey statistics remains challenging. This study a practical effective framework to automatically update existing products bridge gap between remote classification results data. employed Landsat imagery time series, change detection algorithms, sample migration, random forests develop updating China from 1980–2015...

Due to their important role in maintaining temperature and soil moisture, agricultural plastic covers have been widely utilized around the globe for improving crop-growing conditions, which include both plastic-covered greenhouses (PCGs) plastic-mulched farmlands (PMFs). However, it is a challenging long-neglected issue separate PCGs from PMFs due spectral similarity. The objective of this study propose deep semantic segmentation model accurate PCG PMF mapping based on very high-resolution...

An isolation strategy was used to control the transmission and rapid spread of COVID-19 in Yunnan. As a result, students were supposed stay at home disrupted their outside activities. It led detrimental influence on students' mental health. The purpose this study investigate prevalence risk factors depression anxiety among medical provide ideas for prevention students. A cross-sectional survey conducted 2,116 Kunming Medical University from July 8 16, 2020. Participants' demographic living...

The urbanization worldwide leads to the rapid increase of solid waste, posing a threat environment and people's wellbeing. However, it is challenging detect waste sites with high accuracy due complex landscape, very few studies considered mapping across multi-cities in large areas. To tackle this issue, study proposes novel deep learning model for from resolution remote sensing imagery. By integrating multi-scale dilated convolutional neural network (CNN) Swin-Transformer, both local global...

Cropland monitoring is a crucial component for broad user community from Land Use and Cover Change study to food security policy making. Faced with the rich natural ecological environment variable agricultural production conditions of Mid-Spine Belt Beautiful China (MSBBC), this developed novel operational assessment framework that combined near real-time land cover mapping platform (i.e., FROM-GLC Plus), FAO Agricultural Stress Index System, degradation method suggested by United Nations...

Accurate information on the spatial and temporal distribution of abandoned cropland (AC) is crucial for protecting arable land, maintaining regional food security ecological stability. Nevertheless, unavailability dedicated monitoring AC, along with extended time frame required remote sensing surveillance intricate transformation land cover types following abandonment, poses considerable challenges in producing accurate AC maps scientific purposes. To address these challenges, a new...

Change detection, a critical and flourishing Earth observation technology, aims to identify changes through cross-temporal remote sensing images acquired over the same geographical area. With widespread use in various change scenarios, it becomes essential utilize heterogeneous due high challenge of accessing ideal homogeneous images. Nevertheless, domain shift, generated by different imaging factors (e.g., sensors, seasons, atmosphere, illumination), makes unable compare directly. To...

Abstract Textural features of high-resolution remote sensing imagery are a powerful data source for improving classification accuracy because using only spectral information is not sufficient the objects with within-field variability. This study presents methods an object-oriented texture analysis algorithm classification, including wavelet packet transform analysis, grey-level co-occurrence matrix (GLCM) and local spatial statistics. Wavelet method optimization selection feature extraction,...

Abstract A Semivariogram, as defined in geostatistics, is a powerful tool for texture extraction of remotely sensed images. However, the traditional features extracted by semivariogram are generally pixel-based classification. Moreover, most studies have been based on original computation mode and discrete semivariance values. This article describes set (STFs) mean square root pair difference (SRPD) to improve accuracy object-oriented classification (OOC) QuickBird The adaptive parameters...