期刊:
ISPRS Journal of Photogrammetry and Remote Sensing,2023年204:397-420 ISSN:0924-2716
通讯作者:
Meng, R;Zhao, F
作者机构:
[Meng, Ran; Lv, Zhengang; Zhou, Longfei; Meng, R; Huang, Zehua; Xu, Binyuan; Sun, Rui] Huazhong Agr Univ, Coll Resources & Environm, Wuhan 430070, Peoples R China.;[Zhong, Liheng] Ant Grp, Hangzhou 311121, Peoples R China.;[Zhao, Feng] Cent China Normal Univ, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.;[Wu, Jin] Univ Hong Kong, Sch Biol Sci, Hong Kong, Peoples R China.;[Wu, Jin] Chinese Univ Hong Kong, State Key Lab Agrobiotechnol, Shatin, Hong Kong, Peoples R China.
通讯机构:
[Zhao, F ] C;[Meng, R ] H;Huazhong Agr Univ, Coll Resources & Environm, Wuhan 430070, Peoples R China.;Cent China Normal Univ, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.;HIT Artificial Intelligence Res Inst Co Ltd, Harbin, Heilongjiang, Peoples R China.
关键词:
Tree species mapping;Key phenological stage;Transformer;Attention mechanism;Deep learning;Plantation forests
摘要:
Plantation forests provide critical ecosystem services and have experienced worldwide expansion during the past few decades. Accurate mapping of tree species through remote sensing is critical for managing plantation forests. The typical temporal behaviors and traits of tree species in satellite image time series (SITS) generate temporal and spectral features in multiple phenological stages that are critical to improve tree species mapping. However, the diverse input features, sequential relations and complex structures in SITS drastically increase the dimension and difficulty of spectral-temporal feature extraction, which challenges the capacity of many general classifiers not explicitly adapted for spectral-temporal learning. As a result, there is still a lack of a method that could automatically extract spectral-temporal features with high separability and regional adaptability from highdimensional SITS for tree species mapping of plantation forests. Moreover, the effects of varying temporal resolution and feature combination on the plantation tree species mapping are under-explored. Here, we developed a multi-head attention-based method for automatically extracting spectral-temporal features with high separability based on a modified Transformer network (Transformer4SITS) for improved plantation tree species mapping. The end-to-end network model consists of a feature extraction module to learn deep spectral-temporal features from SITS and a fusion module to combine multiple features for improving mapping accuracy. We applied this method to two representative plantation forests in southern and northern China for tree species mapping. The results show that: (1) Transformer4SITS method could self-adaptively extract typical spectraltemporal features of key phenological stages (e.g., greenness rising and falling) from SITS, and achieved significantly improved accuracies by at most 15% in comparison with all four baseline methods (i.e., long shortterm memory, harmonic analysis, time-weighted dynamic time warping, linear discriminant analysis); (2) time series with higher temporal resolution tended to produce more accurate species maps consistently across two sites, with their overall accuracies (OA) respectively increasing from 91.05% and 84.33% (60-day) to 94.88% and 88.72% (5-day), but the effect of high temporal resolution respectively leveled off around 90-day and 50-day resolution across two sites; (3) the mapping results using all available bands and two-band spectral indices outperformed the results using a subset of them, but with only modest increase in the accuracy (i.e., OA increased from 93.63% and 86.01% to 94.88% and 88.72%. This study thus provides a state-of-the-art deep learning-based method for improved tree species mapping, which is critical for sustainable management and biodiversity monitoring of plantation forests across large scales.
作者机构:
[Zhu, Wenchao; Wu, Hao; Jiang, Zhimeng; Cen, Luyu] Cent China Normal Univ, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.;[Zhu, Wenchao; Wu, Hao; Jiang, Zhimeng; Cen, Luyu] Cent China Normal Univ, Hubei Prov Key Lab Geog Proc Anal & Simulat, Wuhan 430079, Peoples R China.
通讯机构:
[Zhimeng Jiang] C;College of Urban and Environmental Sciences, Central China Normal University, Wuhan, China<&wdkj&>Hubei Province Key Laboratory for Geographical Process Analysis and Simulation, Central China Normal University, Wuhan, China
关键词:
land use spatial pattern;resource environment carrying capacity;land use change;spatial optimization;high-quality land development;Zhengzhou city
摘要:
High-intensity land use and resource overloaded-induced regional land use spatial pattern (LUSP) are essential and challenging for high-quality development. The empirical studies have shown that a scientific land uses spatial layout, and the supporting system should be based on a historical perspective and require better considering the double influence between the current characteristics and future dynamics. This study proposes a comprehensive framework that integrates the resource environment carrying capacity (RECC) and land use change (LUC) to investigate strategies for optimizing the spatial pattern of land use for high-quality development. China's Zhengzhou city was the subject of a case study whose datasets include remote sensing, spatial monitoring, statistics, and open sources. Three significant results emerged from the analysis: (1) The RECC has significant spatial differentiation but does not follow a specific spatial law, and regions with relatively perfect ecosystems may not necessarily have better RECC. (2) From 2020 to 2030, the construction land and farmland will fluctuate wildly, with the former increasing by 346.21 km(2) and the latter decreasing by 295.98 km(2). (3) The study area is divided into five zones, including resource conservation, ecological carrying, living core, suitable construction, and grain supply zones, and each one has its LUSP optimization orientation. This uneven distribution of RECC reflects functional defects in the development and utilization of LUSP. In addition, the increase in construction land and the sharp decline of farmland pose potential threats to the sustainable development of the study area. Hence, these two elements cannot be ignored in the future high-quality development process. The findings indicate that the LUSP optimization based on dual dimensions of RECC and LUC is more realistic than a single-dimension solution, exhibiting the LUSP optimization's effectiveness and applicability.
期刊:
IEEE Transactions on Geoscience and Remote Sensing,2023年61:1-14 ISSN:0196-2892
作者机构:
[Xu, Baodong; Zhang, Zhewei; Wei, Haodong; Yang, Jingya; Cai, Zhiwen] Huazhong Agr Univ, Coll Resources & Environm, Macro Agr Res Inst, Wuhan 430070, Peoples R China.;[Wang, Cong] Cent China Normal Univ, Key Lab Geog Proc Anal & Simulat Hubei Prov, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.;[Zhao, Jing; Li, Jing] Chinese Acad Sci, Jointly Sponsored Aerosp Informat Res Inst, State Key Lab Remote Sensing Sci, Beijing 100101, Peoples R China.;[Zhao, Jing; Li, Jing] Beijing Normal Univ, Beijing 100101, Peoples R China.;[Qu, Yonghua] Beijing Normal Univ, Fac Geog Sci, Inst Remote Sensing Sci & Engn, State Key Lab Remote Sensing Sci,Beijing Key Lab, Beijing 100875, Peoples R China.
关键词:
Gaofen satellites;high spatiotemporal resolution;LAINet;leaf area index (LAI);time-series reconstruction
摘要:
High spatiotemporal resolution time series of leaf area index (LAI) are essential for monitoring crop dynamics and validating coarse-resolution LAI products. The optical satellite sensors at decametric resolution have historically suffered from a long revisit cycle and cloud contamination issues that hampered the acquisition of frequent and high-quality observations. The 16-m/four-day resolution of the new-generation Gaofen-1 (GF-1) and Gaofen-6 (GF-6) satellites provide an unprecedented opportunity to address these limitations. Here, we developed an effective strategy to generate daily 16-m LAI maps combining GF-1/6 data and ground LAINet measurements. All high-quality GF-1/6 observations were utilized first to derive smoothed time series of vegetation indices (VIs). Second, a random forest regression (RF-r) model was trained to link the VIs with corresponding field LAI measurements. The trained RF-r was finally employed to generate the LAI maps. Results demonstrated the reliability of the reconstructed daily VIs (relative error (RE) < 1%) and the derived LAI time series, which greatly benefited from GF-1/6 high-frequency observations. The direct comparison with field LAI measurements by LAI-2200/LI-3000 showed the good performance of retrieved LAI maps, with bias, root mean square error (RMSE), and $R^{\mathbf {2}}$ of 0.05, 0.59, and 0.75, respectively. The LAI time series well captured the spatiotemporal variation of crop growth. Furthermore, the continuous GF-1/6 LAI maps outperformed Sentinel-2 LAI estimates both in terms of temporal frequency and accuracy. Our study indicates the potential of GF-1/6 to generate continuous decametric-resolution LAI maps for fine-scale agricultural monitoring.
作者机构:
[Han, Yong; Ni, Ruixing; Deng, Yating] Xinyang Normal Univ, Sch Geog Sci, Xinyang 464000, Peoples R China.;[Zhu, Yuanyuan; Zhu, YY] Cent China Normal Univ, Key Lab Geog Proc Anal & Simulat Hubei Prov, Wuhan 430079, Peoples R China.
通讯机构:
[Zhu, YY ] C;Cent China Normal Univ, Key Lab Geog Proc Anal & Simulat Hubei Prov, Wuhan 430079, Peoples R China.
摘要:
The imbalanced regional development of higher vocational education, particularly the disparity between the supply and demand of educational resources, has emerged as the primary factor impeding the provision of high-quality higher education in China during the establishment of a universal education system. Based on the 1,482 higher vocational education institutions recognized by the Ministry of Education of China in 2021 as the research objects, the development of higher vocational education in China was explored from the perspective of supply and demand using the entropy weight TOPSIS method and coupling coordination degree model. It was found that China's higher vocational institutions were mainly located in provincial capitals, representing a point distribution pattern. From a comprehensive evaluation of the supply level, areas such as the Beijing-Tianjin-Hebei region, Yangtze River Delta, and central Henan Province have become the catchment areas for the development of higher vocational education, laying the foundation for regional network cooperation. From the perspective of educational equality, the higher vocational education in China was found to be sufficient to match the supply and demand, and a balance between supply and demand was apparent in provincial capitals. The coupling degree between supply and demand exhibited an "olive-type" spatial structure pattern, indicating that the development of higher vocational education in most cities in China is still in the transformation stage. The results provide a scientific basis for optimizing resources in the provision of higher vocational education.
作者机构:
[Fan, Junfu; Zuo, Jiwei; Shi, Zongwen; Chen, Jiahao; Zhang, Mengzhen] Shandong Univ Technol, Sch Civil Engn & Geomat, Zibo 255000, Peoples R China.;[Fan, Junfu; Zhang, Mengzhen] Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China.;[Chen, Jiahao] Cent China Normal Univ, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.;[Ji, Min] Shandong Univ Sci & Technol, Coll Geodesy & Geomat, Qingdao 266510, Peoples R China.
通讯机构:
[Chen, JH ] S;Shandong Univ Technol, Sch Civil Engn & Geomat, Zibo 255000, Peoples R China.;Cent China Normal Univ, Coll Urban & Environm Sci, Wuhan 430079, Peoples R China.
关键词:
building change detection;deep learning;high-resolution;multispectral;multisource spectral data
摘要:
Building change detection is an important task in the remote sensing field, and the powerful feature extraction ability of the deep neural network model shows strong advantages in this task. However, the datasets used for this study are mostly three-band high-resolution remote sensing images from a single data source, and few spectral features limit the development of building change detection from multisource remote sensing images. To investigate the influence of spectral and texture features on the effect of building change detection based on deep learning, a multisource building change detection dataset (MS-HS BCD dataset) is produced in this paper using GF-1 high-resolution remote sensing images and Sentinel-2B multispectral remote sensing images. According to the different resolutions of each Sentinel-2B band, eight different multisource spectral data combinations are designed, and six advanced network models are selected for the experiments. After adding multisource spectral and texture feature data, the results show that the detection effects of the six networks improve to different degrees. Taking the MSF-Net network as an example, the F1-score and IOU improved by 0.67% and 1.09%, respectively, compared with high-resolution images, and by 7.57% and 6.21% compared with multispectral images.
摘要:
The tiers of prefectures and counties are important indicators that reflect their political status,population and economic importance.By using historical quantitative analysis and GIS analysis methods,this paper unpacks the spatiotemporal changes of 339 prefectures and ...MORE The tiers of prefectures and counties are important indicators that reflect their political status,population and economic importance.By using historical quantitative analysis and GIS analysis methods,this paper unpacks the spatiotemporal changes of 339 prefectures and 1607 counties in Tang Dynasty.The results show that:(1)The number of prefecture tier Fu(府),Fu(辅),Xiong(雄)and county tier Chi(赤),Ji(畿),Ci-Chi(次赤)and Ci-Ji(次畿)was relatively stable in Tang Dynasty,while the number of Shang(上),Zhong(中),Xia(下)prefectures and counties changed drastically.In the late Tang Dynasty,the number of upgraded prefectures and counties was more than that of degraded prefectures and counties,with the most significant hierarchical change took place from Kaiyuan(713-741)to Yuanhe(806-820).(2)The spatio-temporal changes of prefectures and counties in Tang Dynasty was“high in the north and low in the south”.Guanzhong Plain was the highest area in the prefecture and county level.The temporal change was“falling in the north and rising in the south”.The Plain of Hubei and Hunan,Poyang Lake Plain and Taihu Plain in the middle and lower reaches of the Yangtze River rose most significantly.(3)The tiers of prefectures and counties in the vicinity of the capital of the Tang Dynasty were most affected by political factors,while the tiers of the frontier fortresses and traffic throats were most affected by military factors.Other prefectures and counties tiers were mainly affected by economic factors,especially population size.(4)The spatio-temporal changes of the tiers of prefectures and counties in Tang Dynasty reflected the eastward and southward movement of the national political,demographic,urban and economic centers after the An-Shi Rebellion in the middle of the Tang Dynasty.FEWER
期刊:
European Journal of Agronomy,2023年151 ISSN:1161-0301
通讯作者:
Xu, BD;Hu, Q
作者机构:
[Xu, Baodong; Xu, BD; Zhou, Wei; Cai, Zhiwen] Huazhong Agr Univ, Macro Agr Res Inst, Coll Resources & Environm, Wuhan 430070, Peoples R China.;[Hu, Jie; Zhang, Xinyu; Wei, Haodong; Chen, Yunping] Huazhong Agr Univ, Coll Plant Sci & Technol, Wuhan 430070, Peoples R China.;[Yang, Jingya] Chinese Acad Agr Sci, Inst Agr Resources & Reg Planning, State Key Lab Efficient Utilizat Arid & Semiarid A, Beijing 100081, Peoples R China.;[Hu, Q; Hu, Qiong] Cent China Normal Univ, Coll Urban & Environm Sci, Key Lab Geog Proc Anal & Simulat Hubei Prov, Wuhan 430079, Peoples R China.;[Xiong, Hang] Huazhong Agr Univ, Coll Econ & Management, Wuhan 430070, Peoples R China.
通讯机构:
[Hu, Q ] C;[Xu, BD ] H;Huazhong Agr Univ, Macro Agr Res Inst, Coll Resources & Environm, Wuhan 430070, Peoples R China.;Cent China Normal Univ, Coll Urban & Environm Sci, Key Lab Geog Proc Anal & Simulat Hubei Prov, Wuhan 430079, Peoples R China.
关键词:
Plastic -mulched citrus index;Spectral separability;Intra-annual and interannual analysis;Spatio-temporal variation;Multi -source remote sensing data
摘要:
The technology of canopy plastic mulching has been widely used in citrus orchards for protecting fruit trees from cold damage. Understanding the spatio-temporal dynamics of plastic-mulched citrus (PMC) is of great importance for precision management of citrus orchards. However, monitoring the long-term and large-area PMC dynamics is challenging because the PMC is typically distributed in cloudy and rainy areas with rapid spatial variability, leading to the limited availability of high-quality remotely sensed data. Moreover, it is difficult to collect sufficient field samples in rugged mountainous regions for extracting PMC. To address these limitations, we proposed a new plastic-mulched citrus index (PMCI) based on spectral separability analysis of PMC and other land cover types. The images from Landsat-5/7/8, Sentinel-2, and Gaofen-1 satellites were employed to extract the intra-annual PMC distribution from 2019 to 2020 and interannual PMC distribution from 2008 to 2020 in Yangshuo county, Guangxi Zhuang Autonomous Region, China. Results showed that the PMCI outperformed other widely used indices in PMC extraction with overall accuracy (OA) increased by 0.09-0.3. Besides, the PMCI exhibited good performances in extracting PMC over different observation dates with OA ranging 0.92-0.98 and 0.91-0.98 in the intra-annual and annual PMC maps, respectively. According to the derived PMC time series maps, PMC displayed significant intra-annual spatial variations in the start date and length of plastic mulching period, whereas the interannual variation of PMC revealed long-term technology adoption patterns in the study area. Furthermore, we found that temperature mainly affected intra-annual PMC variability, and that agricultural policy, market factor, neighborhood effect and variety replacement can explain the interannual PMC variability. These results indicate that the developed PMCI can effectively identify PMC over different agricultural regions and observation dates, and the resultant spatio-temporal dynamic PMC maps based on multi-source satellite data provide an important basis for the promotion of PMC at larger scales and the sustainable development of the citrus industry.
关键词:
Ice albedo;data harmonization;spatial window size;validation;arctic and alpine;Google Earth Engine
摘要:
Albedo plays a key role in regulating the absorption of solar radiation within ice surfaces and hence strongly regulates the production of meltwater. A combination of Landsat and Sentinel 2 data provides the longest continuous medium resolution (10-30 m) earth surface observatory records. An albedo product (harmonized satellite albedo, hereafter HSA) has already been developed and validated for the Greenland Ice Sheet (GrIS), using harmonized Landsat 4-8 and Sentinel 2 datasets. In this paper, the HSA was validated for various Arctic and alpine glaciers and ice caps using in situ measurements. We determine the optimal spatial window size in point-to-pixel analysis, the best practices in evaluating remote sensing algorithms with groundtruth data, and cross sensor comparison of the Landsat 9 (L9) and Landsat 8 (L8) data. The impact of the spatial window size on measured ice surface homogeneity and albedo validation was analysed at both local and regional scales. Homogeneity statistics calculated from the grey-level co-occurrence matrix (GLCM) suggest that the ice surface becomes more homogeneous as the image resolution becomes coarser. The optimal spatial window size was found to be 90 m, based on maximizing the statistical and graphical measures while minimizing the root mean square error and bias. HSAs generally agree closely with in situ albedo measurements (e.g. Pearson's R ranges from 0.68 to 0.92) across various Arctic and alpine glaciers and ice caps. Cross sensor differences between L9 and L8 are minor, and we suggest that no harmonization is necessary to add L9 to our HSA product.
期刊:
E3S Web of Conferences,2023年437 ISSN:2267-1242
作者机构:
Institute for Advanced Studies, University of Malaya, 50603 Kuala Lumpur, Malaysia;School of Environment Art, Hubei Institute of Fine Arts, 430205 Wuhan, Hubei, China;College of Urban and Environmental Sciences, Central China Normal University, 430079 Wuhan, Hubei, China
摘要:
Vacant and abandoned spaces were increasingly recognised as a major obstacle to urban revitalisation. How to intervene sustainably in urban void areas became a challenge. This paper presented a new theoretical framework for sustainable operation and design by reflecting and innovating on existing studies through a literature review and classification. The framework for sustainable operation planning included operational theme and promotion, management and maintenance, effective resource utilisation and sustainable digital technology. The framework for sustainable environment design included function and layout, ecosystem management, construction material and art design aesthetic. In the context of the sustainability framework, one urban void area in the northern part of the 19th Middle School in Wuhan was selected as the case study area. This case was based on secondary data collection, field research, virtual 3D modelling and computer rendering. The aim was to achieve sustainable and synergistic economic, social and environmental development by activating urban void areas. The results found that the theoretical framework of sustainability had positive implications for operation and design. The scientific and practical value of the planning and design could be enhanced by fully considering the logic of sustainable operation. In addition, the innovative concepts of the sustainable framework provided a theoretical and practical basis for filling the research gap of interventions in the urban void area.