首页 > 最新文献

International journal of applied earth observation and geoinformation : ITC journal最新文献

英文 中文
An intercomparison of national and global land use and land cover products for Fiji 斐济国家与全球土地利用和土地覆被产品的相互比较
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-17 DOI: 10.1016/j.jag.2024.104260
Kevin P. Davies , John Duncan , Renata Varea , Diana Ralulu , Solomoni Nagaunavou , Nathan Wales , Eleanor Bruce , Bryan Boruff
Here, a methodology to generate national-scale annual 10 m spatial resolution land use and land cover maps for Fiji (Fiji LULC) is presented. A training dataset of 13,419 points with a LULC label across three years from 2019 to 2021 was generated alongside a nationally representative test dataset of 834 points. These data were used to train a random forests model to convert an image stack of pre-processed Sentinel-2 surface reflectance data and topographic spatial layers into an annual categorical LULC map. When evaluated against the test dataset, the model has an overall accuracy of 83 % (SE: 2.1 %).
The Fiji LULC map was compared to three global 10 m spatial resolution land cover products: Google’s Dynamic World, ESRI LULC, and ESA’s WorldCover v200. These maps were compared statistically using the independent test dataset and in several case study applications (e.g. agricultural monitoring and disaster impacts mapping). The Fiji LULC had a higher overall accuracy than the three global LULC products and aligned more closely with a high-quality field survey of over 2500 rice fields (i.e. Fiji LULC classified 88 % of the rice fields as agricultural compared to 60.6–15.7 % in the global LULC products). A comparison of the overlap between the agricultural class of the four LULC maps with a flood mask following Tropical Cyclone Yasa indicated that dataset choice has a substantial impact on estimates of the area of flooded croplands. The Fiji LULC map tends to capture agricultural land covers and smaller scale landscape features with more accuracy than the global products. This analysis illustrates the importance of assessing the performance of global LULC products in particular locations and for specific applications. As demonstrated here, the choice of LULC product could impact subsequent analysis and monitoring tasks. To support these LULC product comparisons, an open-source Python package for computing performance metrics for LULC maps when reference data have different strata to map classes has been published. Further, the training data, test data, and national-scale maps for Fiji have been produced for 2019 to 2022 and are available as open source products on the Pacific Data Hub.
本文介绍了一种生成斐济国家尺度年度 10 米空间分辨率土地利用和土地覆被地图(斐济 LULC)的方法。该方法生成了一个包含 13,419 个点的训练数据集,这些点在 2019 年至 2021 年的三年中带有土地利用、土地覆被和土壤标签,同时还生成了一个包含 834 个点的具有全国代表性的测试数据集。这些数据用于训练一个随机森林模型,将经过预处理的哨兵-2 表面反射率数据和地形空间层的图像堆栈转换为年度分类 LULC 地图。根据测试数据集进行评估后,该模型的总体准确率为 83%(SE:2.1%)。斐济 LULC 地图与三种全球 10 米空间分辨率土地覆被产品进行了比较:斐济 LULC 地图与三种全球 10 米空间分辨率土地覆被产品进行了比较:Google 的 Dynamic World、ESRI LULC 和 ESA 的 WorldCover v200。利用独立测试数据集和几个案例研究应用(如农业监测和灾害影响绘图)对这些地图进行了统计比较。斐济 LULC 的总体准确度高于三种全球 LULC 产品,并且与对 2500 多块稻田进行的高质量实地调查更加吻合(即斐济 LULC 将 88% 的稻田归类为农业用地,而全球 LULC 产品仅将 60.6-15.7% 的稻田归类为农业用地)。在热带气旋 "亚萨 "过后,对四张 LULC 地图的农业类与洪水掩蔽之间的重叠情况进行了比较,结果表明,数据集的选择对洪水淹没耕地面积的估算有很大影响。与全球产品相比,斐济 LULC 地图往往能更准确地捕捉到农业用地覆盖和较小尺度的地貌特征。这一分析表明了评估全球 LULC 产品在特定地点和特定应用中的性能的重要性。正如本文所示,LULC 产品的选择会影响后续的分析和监测任务。为了支持这些 LULC 产品的比较,我们发布了一个开源 Python 软件包,用于计算 LULC 地图的性能指标,当参考数据与地图类别具有不同层级时。此外,斐济 2019 年至 2022 年的培训数据、测试数据和国家尺度地图已经制作完成,并可在太平洋数据枢纽(Pacific Data Hub)上作为开源产品获取。
{"title":"An intercomparison of national and global land use and land cover products for Fiji","authors":"Kevin P. Davies ,&nbsp;John Duncan ,&nbsp;Renata Varea ,&nbsp;Diana Ralulu ,&nbsp;Solomoni Nagaunavou ,&nbsp;Nathan Wales ,&nbsp;Eleanor Bruce ,&nbsp;Bryan Boruff","doi":"10.1016/j.jag.2024.104260","DOIUrl":"10.1016/j.jag.2024.104260","url":null,"abstract":"<div><div>Here, a methodology to generate national-scale annual 10 m spatial resolution land use and land cover maps for Fiji (Fiji LULC) is presented. A training dataset of 13,419 points with a LULC label across three years from 2019 to 2021 was generated alongside a nationally representative test dataset of 834 points. These data were used to train a random forests model to convert an image stack of pre-processed Sentinel-2 surface reflectance data and topographic spatial layers into an annual categorical LULC map. When evaluated against the test dataset, the model has an overall accuracy of 83 % (SE: 2.1 %).</div><div>The Fiji LULC map was compared to three global 10 m spatial resolution land cover products: Google’s Dynamic World, ESRI LULC, and ESA’s WorldCover v200. These maps were compared statistically using the independent test dataset and in several case study applications (e.g. agricultural monitoring and disaster impacts mapping). The Fiji LULC had a higher overall accuracy than the three global LULC products and aligned more closely with a high-quality field survey of over 2500 rice fields (i.e. Fiji LULC classified 88 % of the rice fields as agricultural compared to 60.6–15.7 % in the global LULC products). A comparison of the overlap between the agricultural class of the four LULC maps with a flood mask following Tropical Cyclone Yasa indicated that dataset choice has a substantial impact on estimates of the area of flooded croplands. The Fiji LULC map tends to capture agricultural land covers and smaller scale landscape features with more accuracy than the global products. This analysis illustrates the importance of assessing the performance of global LULC products in particular locations and for specific applications. As demonstrated here, the choice of LULC product could impact subsequent analysis and monitoring tasks. To support these LULC product comparisons, an open-source Python package for computing performance metrics for LULC maps when reference data have different strata to map classes has been published. Further, the training data, test data, and national-scale maps for Fiji have been produced for 2019 to 2022 and are available as open source products on the Pacific Data Hub.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104260"},"PeriodicalIF":7.6,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Combining readily available population and land cover maps to generate non-residential built-up labels to train Sentinel-2 image segmentation models 结合现成的人口和土地覆盖图生成非住宅建筑标签,以训练哨兵-2 图像分割模型
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-17 DOI: 10.1016/j.jag.2024.104272
Diogo Duarte , Cidália C. Fonte
The localization of non-residential buildings over wide geographical areas is used as input within several contexts such as disaster management, regional and national planning, policy making and evaluation, among others. While the built-up environment has been continuously and globally mapped, given the efforts on producing synoptic land cover information; little attention has been given to the land use component of such built-up. This is due to, for example, difficulties in distinguishing built-up land use in non-commercial satellite imagery (e.g., Sentinel-2, with spatial resolution of up to 10 m), difficulties in collecting training data for supervised classification approaches, and the fact that variations in features of the built-up environment not always translate to a specific land use. This is even more critical when considering nadir viewing satellite or aerial imagery. However, map producers have been addressing this issue. For example, the Copernicus program (European Commission), through their pan-European CORINE Land Cover (CLC), and Urban Atlas restricted to several European metropolitan areas, have been making available land use information of the built-up cover, with 6-year intervals. The Global Human Settlement Layer (Copernicus program) has been providing built-up land use information by distinguishing residential from non-residential built-up since 2023 (GHSL_NRES). Currently these are also provided with a time interval of 5 years. National map producers often provide this information but usually with an interval between editions of several years. In this paper we combine readily available population counts and land cover maps to generate non-residential training labels that can be used to train a Sentinel-2 image segmentation model capable of distinguishing non-residential built-up from the remaining built-up. Leveraging two publicly available datasets, population counts (WorldPop) and built-up land cover (ESA WorldCover), allowed to produce training data from which an image segmentation model was able to learn relevant features to distinguish non-residential areas from other built-up in Sentinel-2 images. The results within a study area of 4 Sentinel-2 tiles shown that it improves the detection of non-residential built-up areas when comparing with CLC and GHSL_NRES (F1-score of 32 %, 25 % and 29 %, respectively), which are the products providing pan-European information regarding the built-up land use. These results indicate that the combination of publicly available geospatial datasets may be used to produce higher quality geospatial information.
在灾害管理、区域和国家规划、政策制定和评估等多个方面,对广阔地理区域内的非住宅建筑进行定位都是一种投入。虽然在制作综合土地覆被信息方面做出了努力,对建成区环境进行了持续的全球测绘,但对建成区的土地利用部分却关注甚少。其原因包括:在非商业卫星图像(如空间分辨率高达 10 米的哨兵-2 号)中难以区分建成区的土地利用;难以收集用于监督分类方法的训练数据;以及建成区环境特征的变化并不总能转化为特定的土地利用。当考虑从天顶观测卫星或航空图像时,这一点甚至更为关键。不过,地图制作者一直在解决这个问题。例如,哥白尼计划(欧洲委员会)通过其泛欧洲 CORINE 土地覆盖(CLC)和仅限于欧洲几个大都市地区的城市地图,以 6 年的间隔提供了建筑覆盖的土地利用信息。自 2023 年以来,全球人类住区图层(哥白尼计划)一直在通过区分住宅和非住宅建筑(GHSL_NRES)提供建筑用地信息。目前,这些信息的时间间隔也是 5 年。国家地图编制者通常会提供此类信息,但其版本之间的时间间隔通常为数年。在本文中,我们将现成的人口数量和土地覆被图结合起来,生成非住宅区训练标签,用于训练能够区分非住宅区建筑群和其余建筑群的哨兵-2 图像分割模型。利用两个公开数据集--人口数量(WorldPop)和建成区土地覆盖(ESA WorldCover)--可以生成训练数据,图像分割模型能够从中学习相关特征,以区分哨兵-2 图像中的非居民区和其他建成区。在 4 个哨兵-2 瓦片研究区域内的结果表明,与 CLC 和 GHSL_NRES 相比(F1 分数分别为 32%、25% 和 29%),它提高了对非住宅建筑区的检测能力,而 CLC 和 GHSL_NRES 是提供泛欧建筑用地信息的产品。这些结果表明,将可公开获取的地理空间数据集结合起来,可以生成更高质量的地理空间信息。
{"title":"Combining readily available population and land cover maps to generate non-residential built-up labels to train Sentinel-2 image segmentation models","authors":"Diogo Duarte ,&nbsp;Cidália C. Fonte","doi":"10.1016/j.jag.2024.104272","DOIUrl":"10.1016/j.jag.2024.104272","url":null,"abstract":"<div><div>The localization of non-residential buildings over wide geographical areas is used as input within several contexts such as disaster management, regional and national planning, policy making and evaluation, among others. While the built-up environment has been continuously and globally mapped, given the efforts on producing synoptic land cover information; little attention has been given to the land use component of such built-up. This is due to, for example, difficulties in distinguishing built-up land use in non-commercial satellite imagery (e.g., Sentinel-2, with spatial resolution of up to 10 m), difficulties in collecting training data for supervised classification approaches, and the fact that variations in features of the built-up environment not always translate to a specific land use. This is even more critical when considering nadir viewing satellite or aerial imagery. However, map producers have been addressing this issue. For example, the Copernicus program (European Commission), through their pan-European CORINE Land Cover (CLC), and Urban Atlas restricted to several European metropolitan areas, have been making available land use information of the built-up cover, with 6-year intervals. The Global Human Settlement Layer (Copernicus program) has been providing built-up land use information by distinguishing residential from non-residential built-up since 2023 (GHSL_NRES). Currently these are also provided with a time interval of 5 years. National map producers often provide this information but usually with an interval between editions of several years. In this paper we combine readily available population counts and land cover maps to generate non-residential training labels that can be used to train a Sentinel-2 image segmentation model capable of distinguishing non-residential built-up from the remaining built-up. Leveraging two publicly available datasets, population counts (WorldPop) and built-up land cover (ESA WorldCover), allowed to produce training data from which an image segmentation model was able to learn relevant features to distinguish non-residential areas from other built-up in Sentinel-2 images. The results within a study area of 4 Sentinel-2 tiles shown that it improves the detection of non-residential built-up areas when comparing with CLC and GHSL_NRES (F1-score of 32 %, 25 % and 29 %, respectively), which are the products providing pan-European information regarding the built-up land use. These results indicate that the combination of publicly available geospatial datasets may be used to produce higher quality geospatial information.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104272"},"PeriodicalIF":7.6,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
The illusion of success: Test set disproportion causes inflated accuracy in remote sensing mapping research 成功的假象:测试集比例失调导致遥感测绘研究的精确度膨胀
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-16 DOI: 10.1016/j.jag.2024.104256
Yuanjun Xiao , Zhen Zhao , Jingfeng Huang , Ran Huang , Wei Weng , Gerui Liang , Chang Zhou , Qi Shao , Qiyu Tian
In remote sensing mapping studies, selecting an appropriate test set to accurately evaluate the results is critical. An imprecise accuracy assessment can be misleading and fail to validate the applicability of mapping products. Commencing with the WHU-Hi-HanChuan dataset, this paper revealed the impact of sample size ratios in test sets on accuracy metrics by generating a series of test sets with varying ratios of positive and negative sample size to evaluate the same map. A rigorous approach for accuracy assessment was suggested, and an example of tea plantations mapping is used to demonstrate the process and analyse potential issues in traditional approaches. A scale factor (λ) was constructed to measure the discrepancy in sample size ratios between test sets and actual conditions. Accuracy adjustment formulas were developed and applied to adjust the accuracy of 42 previous maps based on the λ. Results showed a higher ratio of positive to negative sample size in test set led to inflated user’s accuracy (UA), F1-score (F1) and overall accuracy (OA), but had little impact on producer’s accuracy. When the ratio aligned with that in the target area, the UA, F1, and OA closely matched the true values, indicating the proportion of positive and negative samples in test set should be consistent with that in actual situation. The accuracies reported by the traditional approaches including test set sampling from labelled data and 5-fold cross validation were far from the true accuracy and could not reflect the performance of the map. Among 42 previous maps, nearly 60% of the maps had UAs overestimated by 10%, and 9.5% of the maps had UAs and F1s deviations of more than 25%. The conclusions of this study provide a clear caution for future mapping research and assist in producing and identifying truly excellent maps.
在遥感测绘研究中,选择适当的测试集以准确评估结果至关重要。不精确的精度评估可能会产生误导,无法验证测绘产品的适用性。本文从西湖大学-汉川数据集入手,通过生成一系列正负样本量比例不同的测试集来评估同一幅地图,揭示了测试集中样本量比例对精度指标的影响。提出了一种严格的精度评估方法,并以茶园制图为例演示了这一过程,分析了传统方法中可能存在的问题。构建了一个比例因子(λ),用于衡量测试集与实际情况之间样本量比率的差异。根据 λ 制定并应用了精确度调整公式,以调整 42 幅先前地图的精确度。结果显示,测试集中正负样本量的比例越高,用户准确率(UA)、F1 分数(F1)和总体准确率(OA)就越高,但对生产者的准确率影响不大。当比例与目标区域的比例一致时,UA、F1 和 OA 与真实值非常接近,表明测试集中正负样本的比例应与实际情况一致。传统方法(包括从标记数据中抽取测试集样本和 5 倍交叉验证)所报告的准确度与真实准确度相差甚远,无法反映地图的性能。在以往的 42 幅地图中,近 60% 的地图的 UAs 高估了 10%,9.5% 的地图的 UAs 和 F1s 偏差超过 25%。本研究的结论为今后的地图研究提供了明确的警示,有助于制作和识别真正优秀的地图。
{"title":"The illusion of success: Test set disproportion causes inflated accuracy in remote sensing mapping research","authors":"Yuanjun Xiao ,&nbsp;Zhen Zhao ,&nbsp;Jingfeng Huang ,&nbsp;Ran Huang ,&nbsp;Wei Weng ,&nbsp;Gerui Liang ,&nbsp;Chang Zhou ,&nbsp;Qi Shao ,&nbsp;Qiyu Tian","doi":"10.1016/j.jag.2024.104256","DOIUrl":"10.1016/j.jag.2024.104256","url":null,"abstract":"<div><div>In remote sensing mapping studies, selecting an appropriate test set to accurately evaluate the results is critical. An imprecise accuracy assessment can be misleading and fail to validate the applicability of mapping products. Commencing with the WHU-Hi-HanChuan dataset, this paper revealed the impact of sample size ratios in test sets on accuracy metrics by generating a series of test sets with varying ratios of positive and negative sample size to evaluate the same map. A rigorous approach for accuracy assessment was suggested, and an example of tea plantations mapping is used to demonstrate the process and analyse potential issues in traditional approaches. A scale factor (<span><math><mi>λ</mi></math></span>) was constructed to measure the discrepancy in sample size ratios between test sets and actual conditions. Accuracy adjustment formulas were developed and applied to adjust the accuracy of 42 previous maps based on the <span><math><mi>λ</mi></math></span>. Results showed a higher ratio of positive to negative sample size in test set led to inflated user’s accuracy (UA), F1-score (F1) and overall accuracy (OA), but had little impact on producer’s accuracy. When the ratio aligned with that in the target area, the UA, F1, and OA closely matched the true values, indicating the proportion of positive and negative samples in test set should be consistent with that in actual situation. The accuracies reported by the traditional approaches including test set sampling from labelled data and 5-fold cross validation were far from the true accuracy and could not reflect the performance of the map. Among 42 previous maps, nearly 60% of the maps had UAs overestimated by 10%, and 9.5% of the maps had UAs and F1s deviations of more than 25%. The conclusions of this study provide a clear caution for future mapping research and assist in producing and identifying truly excellent maps.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104256"},"PeriodicalIF":7.6,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Multispectral imaging and terrestrial laser scanning for the detection of drought-induced paraheliotropic leaf movement in soybean 多光谱成像和陆地激光扫描用于检测干旱引起的大豆副叶移动
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104250
Erekle Chakhvashvili , Lina Stausberg , Juliane Bendig , Lasse Klingbeil , Bastian Siegmann , Onno Muller , Heiner Kuhlmann , Uwe Rascher
Plant foliage is known to respond rapidly to environmental stressors by adjusting leaf orientation at different timescales. One of the most fascinating mechanisms is paraheliotropism, also known as light avoidance through leaf movement. The leaf orientation (zenith and azimuth angles) is a parameter often overlooked in the plant and remote sensing community due to its challenging measurement procedures under field conditions. In this study, we investigate the synergistic potential of uncrewed aerial vehicle (UAV)-based mutlispectral imaging, terrestrial laser scanning (TLS) and radiative transfer model (RTM) inversion to identify the paraheliotropic response of two distinct soybean varieties: Minngold, a chlorophyll-deficient mutant, and Eiko, a wild variety. We examined their responses to drought stress during the boreal summer drought in 2022 in western Germany by measuring average leaf inclination angle (ALIA) and canopy reflectance. Measurements were taken in the morning and at midday to track leaf movement. Our observations show significant differences between the paraheliotropic response of both varieties. Eiko’s terminal and lateral leaves became vertically erect in the midday (5461), while Minngold’s ALIA remained largely unchanged (5257). Apart from the vertical leaf movement, we also observed leaf inversion (exposing the abaxial side of the leaf) in Eiko under extreme water scarcity. The red edge band at 740 nm showed the strongest correlation with ALIA (r2=0.520.76) The ratio of the far red edge to near infrared (RE740/NIR842) vegetation index compensated for varying light levels during morning and afternoon measurements, exhibiting strong correlations with ALIA when considering only sun-lit leaf spectra (r2=0.72). The retrieval of ALIA with PROSAIL varied based on ALIA constraints and the spectra used for retrieval (full spectrum or the combination of bands 742 and 842), resulting in a root mean square error (RMSE) of 7.7-12.9°. PROSAIL faced challenges in simulating the spectra of plots with very low LAI due to the soil background. This study made the first attempt to observe different paraheliotropic responses of two soybean varieties with UAV-based multispectral imaging. Proximal sensing opens up the possibilities to observe early stress indicators such as paraheliotropism, at much higher spatial and temporal resolution than ever before.
众所周知,植物叶片可通过在不同时间尺度上调整叶片方向,对环境压力做出快速反应。其中最吸引人的机制之一是副向光性,也称为通过叶片运动避光。叶片方向(天顶角和方位角)是植物和遥感界经常忽视的一个参数,因为它在野外条件下的测量程序极具挑战性。在本研究中,我们研究了基于无人机(UAV)的多光谱成像、地面激光扫描(TLS)和辐射传递模型(RTM)反演的协同潜力,以确定两个不同大豆品种的副向光性响应:它们分别是叶绿素缺陷突变体 Minngold 和野生品种 Eiko。我们通过测量平均叶片倾角 (ALIA) 和冠层反射率,研究了这两个品种在 2022 年德国西部北方夏季干旱期间对干旱胁迫的响应。测量在早晨和中午进行,以跟踪叶片的运动。我们的观察结果表明,这两个品种的副向日光反应存在显著差异。英子的顶叶和侧叶在正午时垂直直立(54→61∘),而明戈德的 ALIA 基本保持不变(52→57∘)。除了叶片的垂直移动外,我们还观察到英子在极度缺水的情况下叶片反转(露出叶片背面)。波长为 740 nm 的红边波段与 ALIA 的相关性最强(r2=0.52-0.76)。在上午和下午的测量中,远红边与近红外(RE740/NIR842)植被指数之比补偿了不同的光照水平,当仅考虑阳光照射下的叶片光谱时,与 ALIA 的相关性很强(r2=0.72)。利用 PROSAIL 对 ALIA 的检索因 ALIA 约束条件和用于检索的光谱(全光谱或波段 742 和 842 的组合)而异,导致均方根误差(RMSE)为 7.7-12.9°。由于土壤背景的原因,PROSAIL 在模拟 LAI 很低的地块的光谱时面临挑战。这项研究首次尝试利用基于无人机的多光谱成像技术观测两个大豆品种的不同副向日冕响应。近距离传感技术为观测副向斜等早期胁迫指标提供了可能性,其空间和时间分辨率远远高于以往任何时候。
{"title":"Multispectral imaging and terrestrial laser scanning for the detection of drought-induced paraheliotropic leaf movement in soybean","authors":"Erekle Chakhvashvili ,&nbsp;Lina Stausberg ,&nbsp;Juliane Bendig ,&nbsp;Lasse Klingbeil ,&nbsp;Bastian Siegmann ,&nbsp;Onno Muller ,&nbsp;Heiner Kuhlmann ,&nbsp;Uwe Rascher","doi":"10.1016/j.jag.2024.104250","DOIUrl":"10.1016/j.jag.2024.104250","url":null,"abstract":"<div><div>Plant foliage is known to respond rapidly to environmental stressors by adjusting leaf orientation at different timescales. One of the most fascinating mechanisms is paraheliotropism, also known as light avoidance through leaf movement. The leaf orientation (zenith and azimuth angles) is a parameter often overlooked in the plant and remote sensing community due to its challenging measurement procedures under field conditions. In this study, we investigate the synergistic potential of uncrewed aerial vehicle (UAV)-based mutlispectral imaging, terrestrial laser scanning (TLS) and radiative transfer model (RTM) inversion to identify the paraheliotropic response of two distinct soybean varieties: Minngold, a chlorophyll-deficient mutant, and Eiko, a wild variety. We examined their responses to drought stress during the boreal summer drought in 2022 in western Germany by measuring average leaf inclination angle (ALIA) and canopy reflectance. Measurements were taken in the morning and at midday to track leaf movement. Our observations show significant differences between the paraheliotropic response of both varieties. Eiko’s terminal and lateral leaves became vertically erect in the midday (<span><math><mrow><mn>54</mn><mo>→</mo><mn>6</mn><msup><mrow><mn>1</mn></mrow><mrow><mo>∘</mo></mrow></msup></mrow></math></span>), while Minngold’s ALIA remained largely unchanged (<span><math><mrow><mn>52</mn><mo>→</mo><mn>5</mn><msup><mrow><mn>7</mn></mrow><mrow><mo>∘</mo></mrow></msup></mrow></math></span>). Apart from the vertical leaf movement, we also observed leaf inversion (exposing the abaxial side of the leaf) in Eiko under extreme water scarcity. The red edge band at 740 nm showed the strongest correlation with ALIA (<span><math><mrow><msup><mrow><mi>r</mi></mrow><mrow><mn>2</mn></mrow></msup><mo>=</mo><mn>0</mn><mo>.</mo><mn>52</mn><mo>−</mo><mn>0</mn><mo>.</mo><mn>76</mn></mrow></math></span>) The ratio of the far red edge to near infrared (RE740/NIR842) vegetation index compensated for varying light levels during morning and afternoon measurements, exhibiting strong correlations with ALIA when considering only sun-lit leaf spectra (<span><math><mrow><msup><mrow><mi>r</mi></mrow><mrow><mn>2</mn></mrow></msup><mo>=</mo><mn>0</mn><mo>.</mo><mn>72</mn></mrow></math></span>). The retrieval of ALIA with PROSAIL varied based on ALIA constraints and the spectra used for retrieval (full spectrum or the combination of bands 742 and 842), resulting in a root mean square error (RMSE) of 7.7-12.9°. PROSAIL faced challenges in simulating the spectra of plots with very low LAI due to the soil background. This study made the first attempt to observe different paraheliotropic responses of two soybean varieties with UAV-based multispectral imaging. Proximal sensing opens up the possibilities to observe early stress indicators such as paraheliotropism, at much higher spatial and temporal resolution than ever before.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104250"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651864","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Tracking gain and loss of impervious surfaces by integrating continuous change detection and multitemporal classifications from 1985 to 2022 in Beijing 通过整合 1985 年至 2022 年连续变化检测和多时空分类,跟踪北京不透水地表的增减情况
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104268
Xiao Zhang , Liangyun Liu , Wenhan Zhang , Linlin Guan , Ming Bai , Tingting Zhao , Zhehua Li , Xidong Chen
Impervious surfaces are important indicators of human activity, and finding ways to quantify the gain and loss of impervious surfaces is important for sustainable urban development. However, most relevant studies assume that the transformation of natural surfaces to impervious surfaces is irreversible; thus, the losses of impervious surfaces are often ignored. Here, we propose a novel framework taking advantage of continuous change detection, multitemporal classification, and LandTrendr optimization to track the annual gains and losses in impervious surfaces. It may be the first study to focus on both loss and gain of impervious surfaces using time-series Landsat imagery. Specifically, we built dual continuous-change-detection models to pursue lower commission and omission errors for generating time-series training samples. Then, we adopted time-series classifications from multisource information and derived training samples to develop annual impervious-surface maps from 1985 to 2022 in Beijing. Afterwards, a novel optimization algorithm considering spatial heterogeneity and taking advantage of the LandTrendr algorithm was also proposed to optimize the spatiotemporal consistency of these impervious-surface maps. We further calculated accuracy metrics for the proposed method using time-series validation points, finding overall accuracies of 92.91 %±0.97 % and 93.17 %±1.26 % for gains and losses in impervious surfaces, respectively, using a one-year tolerance. Lastly, we revealed the gains and losses of impervious surfaces in Beijing during 1985–2022. The gained area of impervious surfaces was found to be 1996.21 km2 ± 18.58 km2, and there was a rapid increase during 2000–2010; the total lost area of impervious surfaces was 898.60 km2 ± 4.58 km2, of which 564.85 km2 ± 2.21 km2 first increased and was then lost. Therefore, the proposed method provides a new way of tracking the gain and loss of impervious surfaces, and it offers new possibilities for monitoring urban regreening.
不透水表面是人类活动的重要指标,找到量化不透水表面增减的方法对城市可持续发展非常重要。然而,大多数相关研究都假定自然表面向不透水表面的转化是不可逆的,因此,不透水表面的损失往往被忽视。在此,我们提出了一个新颖的框架,利用连续变化检测、多时相分类和 LandTrendr 优化来跟踪不透水表面的年度增减。这可能是第一项利用时间序列大地遥感卫星图像同时关注不透水表面损耗和增加的研究。具体来说,我们建立了双重连续变化检测模型,以追求在生成时间序列训练样本时降低委托误差和遗漏误差。然后,我们采用来自多源信息的时间序列分类和得出的训练样本,绘制了北京从 1985 年到 2022 年的年度不透水面地图。之后,我们还提出了一种考虑空间异质性并利用 LandTrendr 算法的新型优化算法,以优化这些不透水面地图的时空一致性。我们利用时间序列验证点进一步计算了所提方法的精度指标,发现在一年容差范围内,不透水面增减的总体精度分别为 92.91 %±0.97 % 和 93.17 %±1.26 %。最后,我们揭示了 1985-2022 年间北京不透水地面的增减情况。结果发现,不透水地面的增加面积为 1996.21 平方公里±18.58 平方公里,并且在 2000-2010 年期间出现了快速增长;不透水地面的总损失面积为 898.60 平方公里±4.58 平方公里,其中 564.85 平方公里±2.21 平方公里先增加后损失。因此,所提出的方法为跟踪不透水地面的增加和减少提供了一种新的途径,也为监测城市绿化提供了新的可能性。
{"title":"Tracking gain and loss of impervious surfaces by integrating continuous change detection and multitemporal classifications from 1985 to 2022 in Beijing","authors":"Xiao Zhang ,&nbsp;Liangyun Liu ,&nbsp;Wenhan Zhang ,&nbsp;Linlin Guan ,&nbsp;Ming Bai ,&nbsp;Tingting Zhao ,&nbsp;Zhehua Li ,&nbsp;Xidong Chen","doi":"10.1016/j.jag.2024.104268","DOIUrl":"10.1016/j.jag.2024.104268","url":null,"abstract":"<div><div>Impervious surfaces are important indicators of human activity, and finding ways to quantify the gain and loss of impervious surfaces is important for sustainable urban development. However, most relevant studies assume that the transformation of natural surfaces to impervious surfaces is irreversible; thus, the losses of impervious surfaces are often ignored. Here, we propose a novel framework taking advantage of continuous change detection, multitemporal classification, and LandTrendr optimization to track the annual gains and losses in impervious surfaces. It may be the first study to focus on both loss and gain of impervious surfaces using time-series Landsat imagery. Specifically, we built dual continuous-change-detection models to pursue lower commission and omission errors for generating time-series training samples. Then, we adopted time-series classifications from multisource information and derived training samples to develop annual impervious-surface maps from 1985 to 2022 in Beijing. Afterwards, a novel optimization algorithm considering spatial heterogeneity and taking advantage of the LandTrendr algorithm was also proposed to optimize the spatiotemporal consistency of these impervious-surface maps. We further calculated accuracy metrics for the proposed method using time-series validation points, finding overall accuracies of 92.91 %±0.97 % and 93.17 %±1.26 % for gains and losses in impervious surfaces, respectively, using a one-year tolerance. Lastly, we revealed the gains and losses of impervious surfaces in Beijing during 1985–2022. The gained area of impervious surfaces was found to be 1996.21 km<sup>2</sup> ± 18.58 km<sup>2</sup>, and there was a rapid increase during 2000–2010; the total lost area of impervious surfaces was 898.60 km<sup>2</sup> ± 4.58 km<sup>2</sup>, of which 564.85 km<sup>2</sup> ± 2.21 km<sup>2</sup> first increased and was then lost. Therefore, the proposed method provides a new way of tracking the gain and loss of impervious surfaces, and it offers new possibilities for monitoring urban regreening.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104268"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
White blanket, blue waters: Tracing El Niño footprints in Canada 白色的毯子,蓝色的海水:追寻厄尔尼诺现象在加拿大的足迹
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104267
Afshin Amiri , Silvio Gumiere , Hossein Bonakdari
The El Niño Southern Oscillation (ENSO) significantly influences global climate patterns, with one of the strongest warm phases (El Niño) occurring in 2023, altering precipitation and temperature regimes. In this study, the spatiotemporal variability in snow cover across Canadian provinces from December 2023 to February 2024 relative to long-term averages is explored. The NOAA-OISST, NOAA-CSFV2, and MODIS MOD10A1 remote sensing datasets were selected to assess the impacts of El Niño on snow cover changes and the subsequent effects on water availability, agricultural productivity, the municipal water supply, natural ecosystems, and wildfire risk in Canada. An analysis of sea surface temperature anomalies in the equatorial Pacific revealed that El Niño intensity and progression are linked to regional snow cover deviations. Compared with the long-term average, Canada’s snow cover area experienced significant declines in December 2023, January 2024, and February 2024, with decreases of 135,938 km2 (−7.43 %), 309,928 km2 (−15.26 %), and 136,406 km2 (−4.57 %), respectively. The findings indicate significant disparities among provinces, with Ontario, Quebec, and Manitoba experiencing marked decreases in snow cover, whereas in Saskatchewan and Alberta, initial increases were followed by subsequent variability. In British Columbia, a late-season increase in snow was observed, whereas minor changes were noted in the Maritime provinces and Northern territories. The findings of this study highlight the importance of snow cover as an important factor that has a considerable impact on the hydrological cycle and agricultural productivity, influences environmental health and economic resilience, and is crucial for both natural ecosystems and human livelihoods.
厄尔尼诺南方涛动(ENSO)对全球气候模式有重大影响,其中最强的暖相(厄尔尼诺)出现在 2023 年,改变了降水和温度机制。本研究探讨了 2023 年 12 月至 2024 年 2 月加拿大各省积雪相对于长期平均值的时空变化。研究选择了 NOAA-OISST、NOAA-CSFV2 和 MODIS MOD10A1 遥感数据集,以评估厄尔尼诺现象对积雪覆盖变化的影响,以及随后对加拿大水供应、农业生产力、市政供水、自然生态系统和野火风险的影响。对赤道太平洋海面温度异常的分析表明,厄尔尼诺现象的强度和发展与区域雪盖偏差有关。与长期平均值相比,加拿大的积雪面积在 2023 年 12 月、2024 年 1 月和 2024 年 2 月出现大幅下降,分别减少了 135,938 平方公里(-7.43%)、309,928 平方公里(-15.26%)和 136,406 平方公里(-4.57%)。研究结果表明,各省之间的差异很大,安大略省、魁北克省和马尼托巴省的积雪面积明显减少,而萨斯喀彻温省和阿尔伯塔省的积雪面积在最初增加后随之出现变化。在不列颠哥伦比亚省,观察到晚季积雪增加,而在海洋省份和北部地区则发现了轻微的变化。这项研究的结果凸显了积雪作为一个重要因素的重要性,它对水文循环和农业生产率有相当大的影响,影响环境健康和经济恢复能力,对自然生态系统和人类生计都至关重要。
{"title":"White blanket, blue waters: Tracing El Niño footprints in Canada","authors":"Afshin Amiri ,&nbsp;Silvio Gumiere ,&nbsp;Hossein Bonakdari","doi":"10.1016/j.jag.2024.104267","DOIUrl":"10.1016/j.jag.2024.104267","url":null,"abstract":"<div><div>The El Niño Southern Oscillation (ENSO) significantly influences global climate patterns, with one of the strongest warm phases (El Niño) occurring in 2023, altering precipitation and temperature regimes. In this study, the spatiotemporal variability in snow cover across Canadian provinces from December 2023 to February 2024 relative to long-term averages is explored. The NOAA-OISST, NOAA-CSFV2, and MODIS MOD10A1 remote sensing datasets were selected to assess the impacts of El Niño on snow cover changes and the subsequent effects on water availability, agricultural productivity, the municipal water supply, natural ecosystems, and wildfire risk in Canada. An analysis of sea surface temperature anomalies in the equatorial Pacific revealed that El Niño intensity and progression are linked to regional snow cover deviations. Compared with the long-term average, Canada’s snow cover area experienced significant declines in December 2023, January 2024, and February 2024, with decreases of 135,938 km<sup>2</sup> (−7.43 %), 309,928 km<sup>2</sup> (−15.26 %), and 136,406 km<sup>2</sup> (−4.57 %), respectively. The findings indicate significant disparities among provinces, with Ontario, Quebec, and Manitoba experiencing marked decreases in snow cover, whereas in Saskatchewan and Alberta, initial increases were followed by subsequent variability. In British Columbia, a late-season increase in snow was observed, whereas minor changes were noted in the Maritime provinces and Northern territories. The findings of this study highlight the importance of snow cover as an important factor that has a considerable impact on the hydrological cycle and agricultural productivity, influences environmental health and economic resilience, and is crucial for both natural ecosystems and human livelihoods.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104267"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
DeLA: An extremely faster network with decoupled local aggregation for large scale point cloud learning DeLA:用于大规模点云学习的具有解耦局部聚合功能的极速网络
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104255
Weikang Yang , Xinghao Lu , Binjie Chen , Chenlu Lin , Xueye Bao , Weiquan Liu , Yu Zang , Junyu Xu , Cheng Wang
With advances in data collection technology, the volume of recent remote sensing point cloud datasets has grown significantly, posing substantial challenges for point cloud deep learning, particularly in neighborhood aggregation operations. Unlike simple pooling, neighborhood aggregation incorporates spatial relationships between points into the feature aggregation process, requiring repeated relationship learning and resulting in substantial computational redundancy. The exponential increase in data volume exacerbates this issue. To address this, we theoretically demonstrate that if basic spatial information is encoded in point features, simple pooling operations can effectively aggregate features. This means the spatial relationships can be extracted and integrated with other features during aggregation. Based on this concept, we propose a lightweight point network called DeLA (Decoupled Local Aggregation). DeLA separates the traditional neighborhood aggregation process into distinct spatial encoding and local aggregation operations, reducing the computational complexity by a factor of K, where K is the number of neighbors in the K-Nearest Neighbor algorithm (K-NN). Experimental results on five classic benchmarks show that DeLA achieves state-of-the-art performance with reduced or equivalent latency. Specifically, DeLA exceeds 90% overall accuracy on ScanObjectNN and 74% mIoU on S3DIS Area 5. Additionally, DeLA achieves state-of-the-art results on ScanNetV2 with only 20% of the parameters of equivalent models. Our code is available at https://github.com/Matrix-ASC/DeLA.
随着数据采集技术的进步,近期遥感点云数据集的数量大幅增加,这给点云深度学习,尤其是邻域聚合操作带来了巨大挑战。与简单的池化不同,邻域聚合将点与点之间的空间关系纳入特征聚合过程,需要反复学习关系,造成大量计算冗余。数据量的指数级增长加剧了这一问题。为了解决这个问题,我们从理论上证明,如果在点特征中编码了基本空间信息,那么简单的汇集操作就能有效地聚合特征。这意味着在聚合过程中,可以提取空间关系并与其他特征进行整合。基于这一概念,我们提出了一种名为 DeLA(解耦局部聚合)的轻量级点网络。DeLA 将传统的邻域聚合过程分离为不同的空间编码和局部聚合操作,将计算复杂度降低了 K 倍,其中 K 是 K-NN 算法(K-Nearest Neighbor algorithm)中的邻域数。在五个经典基准上的实验结果表明,DeLA 在降低或等同延迟的情况下实现了最先进的性能。具体来说,DeLA 在 ScanObjectNN 上的总体准确率超过 90%,在 S3DIS Area 5 上的 mIoU 超过 74%。此外,DeLA 在 ScanNetV2 上取得了最先进的结果,其参数仅为同等模型的 20%。我们的代码见 https://github.com/Matrix-ASC/DeLA。
{"title":"DeLA: An extremely faster network with decoupled local aggregation for large scale point cloud learning","authors":"Weikang Yang ,&nbsp;Xinghao Lu ,&nbsp;Binjie Chen ,&nbsp;Chenlu Lin ,&nbsp;Xueye Bao ,&nbsp;Weiquan Liu ,&nbsp;Yu Zang ,&nbsp;Junyu Xu ,&nbsp;Cheng Wang","doi":"10.1016/j.jag.2024.104255","DOIUrl":"10.1016/j.jag.2024.104255","url":null,"abstract":"<div><div>With advances in data collection technology, the volume of recent remote sensing point cloud datasets has grown significantly, posing substantial challenges for point cloud deep learning, particularly in neighborhood aggregation operations. Unlike simple pooling, neighborhood aggregation incorporates spatial relationships between points into the feature aggregation process, requiring repeated relationship learning and resulting in substantial computational redundancy. The exponential increase in data volume exacerbates this issue. To address this, we theoretically demonstrate that if basic spatial information is encoded in point features, simple pooling operations can effectively aggregate features. This means the spatial relationships can be extracted and integrated with other features during aggregation. Based on this concept, we propose a lightweight point network called DeLA (Decoupled Local Aggregation). DeLA separates the traditional neighborhood aggregation process into distinct spatial encoding and local aggregation operations, reducing the computational complexity by a factor of K, where K is the number of neighbors in the K-Nearest Neighbor algorithm (K-NN). Experimental results on five classic benchmarks show that DeLA achieves state-of-the-art performance with reduced or equivalent latency. Specifically, DeLA exceeds 90% overall accuracy on ScanObjectNN and 74% mIoU on S3DIS Area 5. Additionally, DeLA achieves state-of-the-art results on ScanNetV2 with only 20% of the parameters of equivalent models. Our code is available at <span><span>https://github.com/Matrix-ASC/DeLA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104255"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Using UAV hyperspectral imagery and deep learning for Object-Based quantitative inversion of Zanthoxylum rust disease index 利用无人机高光谱成像和深度学习,实现基于对象的黄腐病锈病指数定量反演
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104262
Kai Zhang , Jie Deng , Congying Zhou , Jiangui Liu , Xuan Lv , Ying Wang , Enhong Sun , Yan Liu , Zhanhong Ma , Jiali Shang
Zanthoxylum rust (ZR) poses a significant threat to Zanthoxylum bungeanum Maxim.(ZBM) production, impacting both the yield and quality. The lack of current research on ZR using unmanned aerial vehicle (UAV) remote sensing poses a challenge to achieving precise management of individual ZBM plant. This study acquired six UAV hyperspectral images to create a ZR inversion dataset . This dataset, to our knowledge, is the first dataset for remote sensing deep learning (DL) of ZR using UAV. To facilitate automated extraction of individual ZBM plant and the quantitative inversion of ZR disease index (DI), we introduced the object-based quantitative inversion framework (OQIF). OQIF achieved high accuracy in recognizing ZBM (average precision at an intersection over union threshold of 0.5 was 90.0 %). Remarkably, OQIF demonstrates outstanding quantitative inversion results for ZR DI (R2 = 0.90, RMSE = 3.97, n = 8166). For DI < 10, the RMSE was 2.48, showcasing early detection capability. Our research has significant implications for ZBM cultivation and precision management, pioneering object-based quantitative inversion for tree diseases and yield estimation, with potential for early ZR detection.
Zanthoxylum rust(ZR)对Zanthoxylum bungeanum Maxim.(ZBM)的产量和质量都构成了严重威胁。目前缺乏利用无人飞行器(UAV)遥感技术对 ZR 进行研究,这对实现对单株 ZBM 植物的精确管理构成了挑战。这项研究获取了六幅无人机高光谱图像,以创建 ZR 反演数据集。据我们所知,该数据集是首个利用无人机进行 ZR 遥感深度学习(DL)的数据集。为了便于自动提取单株 ZBM 植物和定量反演 ZR 疾病指数(DI),我们引入了基于对象的定量反演框架(OQIF)。OQIF 在识别 ZBM 方面达到了很高的精确度(交叉点超过结合阈值 0.5 时的平均精确度为 90.0%)。值得注意的是,OQIF 对 ZR DI 的定量反演结果非常出色(R2 = 0.90,RMSE = 3.97,n = 8166)。对于 DI < 10,RMSE 为 2.48,显示了早期检测能力。我们的研究对 ZBM 栽培和精确管理具有重要意义,开创了基于对象的树木病害和产量估算定量反演,具有早期 ZR 检测的潜力。
{"title":"Using UAV hyperspectral imagery and deep learning for Object-Based quantitative inversion of Zanthoxylum rust disease index","authors":"Kai Zhang ,&nbsp;Jie Deng ,&nbsp;Congying Zhou ,&nbsp;Jiangui Liu ,&nbsp;Xuan Lv ,&nbsp;Ying Wang ,&nbsp;Enhong Sun ,&nbsp;Yan Liu ,&nbsp;Zhanhong Ma ,&nbsp;Jiali Shang","doi":"10.1016/j.jag.2024.104262","DOIUrl":"10.1016/j.jag.2024.104262","url":null,"abstract":"<div><div><em>Zanthoxylum</em> rust (ZR) poses a significant threat to <em>Zanthoxylum bungeanum</em> Maxim.(ZBM) production, impacting both the yield and quality. The lack of current research on ZR using unmanned aerial vehicle (UAV) remote sensing poses a challenge to achieving precise management of individual ZBM plant. This study acquired six UAV hyperspectral images to create a ZR inversion dataset . This dataset, to our knowledge, is the first dataset for remote sensing deep learning (DL) of ZR using UAV. To facilitate automated extraction of individual ZBM plant and the quantitative inversion of ZR disease index (DI), we introduced the object-based quantitative inversion framework (OQIF). OQIF achieved high accuracy in recognizing ZBM (average precision at an intersection over union threshold of 0.5 was 90.0 %). Remarkably, OQIF demonstrates outstanding quantitative inversion results for ZR DI (R<sup>2</sup> = 0.90, RMSE = 3.97, n = 8166). For DI &lt; 10, the RMSE was 2.48, showcasing early detection capability. Our research has significant implications for ZBM cultivation and precision management, pioneering object-based quantitative inversion for tree diseases and yield estimation, with potential for early ZR detection.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104262"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Accuracy fluctuations of ICESat-2 height measurements in time series ICESat-2 时间序列高度测量的精度波动
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104234
Xu Wang , Xinlian Liang , Weishu Gong , Pasi Häkli , Yunsheng Wang
The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) mission, spanning the past five years, has collected extensive three-dimensional Earth observation data, facilitating the understanding of environmental changes on a global scale. Its key product, Land and Vegetation Height (ATL08), offers global land and vegetation height data for carbon budget and cycle modeling. Consistent measurement accuracy of ATL08 is crucial for reliable time series analysis. However, fluctuations in the temporal accuracy of ATL08 data have been ignored in previous studies, leading to unknown uncertainties in existing time-series analyses. To bridge the knowledge gap, this study analyzes 59 months of ATL08 version 006 data in Finland to assess terrain and surface height accuracy, with a focus on temporal fluctuations across six major land cover types. A random forest (RF) model is employed to quantify the relative importance of error factors affecting height accuracy. Moreover, the study assesses accuracy at two official spatial resolutions, i.e., 100 m × 11 m and 20 m × 11 m, to evaluate the capability of ATL08 for the high-resolution height retrieval. For the terrain, the 100 m segment shows a bias of 0.04 m, a mean absolute error (MAE) of 0.44 m, and a root mean square error (RMSE) of 0.66 m, while the 20 m segment exhibits a bias of 0.10 m, a MAE of 0.35 m, and an RMSE of 0.49 m. For the surface height, the 100 m segment shows a bias of −0.59 m, a MAE of 3.06 m, an RMSE of 4.52 m, a bias% of −3.45 %, a MAE% of 21.26 %, and an RMSE% of 31.40 %. The 20 m segment exhibits a bias of −0.72 m, a MAE of 3.51 m, an RMSE of 5.23 m, a bias% of −5.81 %, a MAE% of 28.52 %, and an RMSE% of 42.47 %. The results indicate that improving segment resolution enhances terrain accuracy but reduces surface height accuracy. According to the error factor analysis, surface coverage and beam type are crucial for terrain retrieval accuracy, with their effects varying over time. Seasonal changes, particularly the presence of snow, affect terrain retrieval accuracy, with the lowest accuracy observed around March each year. This study confirms the critical impact of surface height on its retrieval accuracy and suggests avoiding the use of ATL08 for retrieving low target surface heights, especially in steep terrains. Nevertheless, the analysis affirms the applicability of ATL08 for canopy height estimation in boreal forests, primarily composed of coniferous species, highlighting its potential for extensive spatial and temporal research. This contributes to bridging the gaps between accurate estimates and large area coverage in global carbon budget and cycle studies. Additionally, the findings reveal that similar issues may exist in other satellite laser altimetry missions, emphasizing the important impacts of temporal fluctuations in surface and terrain accuracy when utilizing satellite laser altimetry datasets.
冰、云和陆地高程卫星-2(ICESat-2)飞行任务在过去五年中收集了大量三维地球观测数据,促进了对全球环境变化的了解。其主要产品陆地和植被高度(ATL08)为碳预算和碳循环建模提供了全球陆地和植被高度数据。ATL08 测量精度的一致性对于可靠的时间序列分析至关重要。然而,以往的研究忽略了 ATL08 数据时间精度的波动,导致现有的时间序列分析存在未知的不确定性。为弥补这一知识空白,本研究分析了芬兰 59 个月的 ATL08 006 版数据,以评估地形和地表高度精度,重点关注六种主要土地覆被类型的时间波动。采用随机森林 (RF) 模型来量化影响高度精度的误差因素的相对重要性。此外,研究还评估了两种官方空间分辨率(即 100 米×11 米和 20 米×11 米)下的精度,以评估 ATL08 在高分辨率高度检索方面的能力。在地形方面,100 米分辨率段的偏差为 0.04 米,平均绝对误差(MAE)为 0.44 米,均方根误差(RMSE)为 0.66 米;20 米分辨率段的偏差为 0.10 米,MAE 为 0.在地表高度方面,100 米航段的偏差为-0.59 米,均方根误差为 3.06 米,均方根误差为 4.52 米,偏差%为-3.45%,均方根误差%为 21.26%,均方根误差%为 31.40%。20 米分段的偏差为-0.72 米,最大允许误差为 3.51 米,有效误差为 5.23 米,偏差%为-5.81%,最大允许误差为 28.52%,有效误差为 42.47%。结果表明,提高分段分辨率可提高地形精度,但会降低地表高度精度。根据误差因素分析,地表覆盖率和光束类型对地形检索精度至关重要,其影响随时间而变化。季节变化,尤其是下雪,会影响地形检索精度,每年 3 月前后的精度最低。这项研究证实了地表高度对检索精度的关键影响,并建议避免使用 ATL08 来检索低目标地表高度,尤其是在陡峭的地形中。尽管如此,分析结果还是肯定了 ATL08 在北方森林(主要由针叶树种组成)冠层高度估算中的适用性,突出了其在广泛的时空研究中的潜力。这有助于缩小全球碳预算和碳循环研究中精确估算和大面积覆盖之间的差距。此外,研究结果还揭示了其他卫星激光测高任务中可能存在的类似问题,强调了在利用卫星激光测高数据集时,地表和地形精度的时间波动所产生的重要影响。
{"title":"Accuracy fluctuations of ICESat-2 height measurements in time series","authors":"Xu Wang ,&nbsp;Xinlian Liang ,&nbsp;Weishu Gong ,&nbsp;Pasi Häkli ,&nbsp;Yunsheng Wang","doi":"10.1016/j.jag.2024.104234","DOIUrl":"10.1016/j.jag.2024.104234","url":null,"abstract":"<div><div>The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) mission, spanning the past five years, has collected extensive three-dimensional Earth observation data, facilitating the understanding of environmental changes on a global scale. Its key product, Land and Vegetation Height (ATL08), offers global land and vegetation height data for carbon budget and cycle modeling. Consistent measurement accuracy of ATL08 is crucial for reliable time series analysis. However, fluctuations in the temporal accuracy of ATL08 data have been ignored in previous studies, leading to unknown uncertainties in existing time-series analyses. To bridge the knowledge gap, this study analyzes 59 months of ATL08 version 006 data in Finland to assess terrain and surface height accuracy, with a focus on temporal fluctuations across six major land cover types. A random forest (RF) model is employed to quantify the relative importance of error factors affecting height accuracy. Moreover, the study assesses accuracy at two official spatial resolutions, i.e., 100 m × 11 m and 20 m × 11 m, to evaluate the capability of ATL08 for the high-resolution height retrieval. For the terrain, the 100 m segment shows a bias of 0.04 m, a mean absolute error (MAE) of 0.44 m, and a root mean square error (RMSE) of 0.66 m, while the 20 m segment exhibits a bias of 0.10 m, a MAE of 0.35 m, and an RMSE of 0.49 m. For the surface height, the 100 m segment shows a bias of −0.59 m, a MAE of 3.06 m, an RMSE of 4.52 m, a bias% of −3.45 %, a MAE% of 21.26 %, and an RMSE% of 31.40 %. The 20 m segment exhibits a bias of −0.72 m, a MAE of 3.51 m, an RMSE of 5.23 m, a bias% of −5.81 %, a MAE% of 28.52 %, and an RMSE% of 42.47 %. The results indicate that improving segment resolution enhances terrain accuracy but reduces surface height accuracy. According to the error factor analysis, surface coverage and beam type are crucial for terrain retrieval accuracy, with their effects varying over time. Seasonal changes, particularly the presence of snow, affect terrain retrieval accuracy, with the lowest accuracy observed around March each year. This study confirms the critical impact of surface height on its retrieval accuracy and suggests avoiding the use of ATL08 for retrieving low target surface heights, especially in steep terrains. Nevertheless, the analysis affirms the applicability of ATL08 for canopy height estimation in boreal forests, primarily composed of coniferous species, highlighting its potential for extensive spatial and temporal research. This contributes to bridging the gaps between accurate estimates and large area coverage in global carbon budget and cycle studies. Additionally, the findings reveal that similar issues may exist in other satellite laser altimetry missions, emphasizing the important impacts of temporal fluctuations in surface and terrain accuracy when utilizing satellite laser altimetry datasets.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104234"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A high temporal resolution NDVI time series to monitor drought events in the Horn of Africa 监测非洲之角干旱事件的高时间分辨率 NDVI 时间序列
IF 7.6 Q1 REMOTE SENSING Pub Date : 2024-11-15 DOI: 10.1016/j.jag.2024.104264
Riccardo D’Ercole , Daniele Casella , Giulia Panegrossi , Paolo Sanò
This study investigates the reconstruction of climatological patterns and vegetation dynamics in the Horn of Africa region using high temporal resolution (i.e. daily) Normalized Difference Vegetation Index (NDVI) datasets. The analysis compares a straight-forward processing approach to derive a daily vegetation index from a geostationary (SEVIRI) satellite with existing NDVI series from geostationary or polar-orbiting (MODIS, MetOp-AVHRR) satellites, highlighting the impact of cloud contamination on data quality in high temporal resolution datasets. Using a smoothing process designed to reconstruct the upper envelope of the vegetation status series, we obtained a daily vegetation dataset that effectively mitigated cloud-induced fluctuations, outperforming polar-orbiting (e.g. MODIS) satellite-derived dataset in capturing regional climatology. We demonstrated this through statistical analysis, including autocorrelation and mean absolute difference between consecutive observations. We showed that cloud contamination significantly affects high temporal resolution NDVI series, particularly in forest areas, which makes it difficult to identify a suitable dataset to validate our approach. Therefore, we mitigated this problem using a Maximum Value Compositing technique, designed to remove cloud-induced biases and further compared our results with another independent vegetation index at coarser temporal resolution derived from AVHRR. We found that our vegetation index closely relates with MODIS 10-day composites after removing cloud-contaminated pixels. Furthermore, the study evaluates the sensitivity of the selected NDVI datasets to drought events, demonstrating the strength of the proposed SEVIRI dataset in capturing the intensity and persistence of vegetation anomalies. In conclusion, the study presents an innovative strategy for deriving daily-resolution NDVI datasets in cloud-prone regions, validating it with independent datasets at different sub-monthly temporal scales.
本研究利用高时间分辨率(即每日)归一化差异植被指数数据集,调查非洲之角地区气候模式和植被动态的重建情况。分析比较了从地球静止卫星(SEVIRI)和现有的地球静止或极轨道卫星(MODIS、MetOp-AVHRR)归一化差异植被指数系列中得出每日植被指数的简单处理方法,突出了云污染对高时间分辨率数据集数据质量的影响。通过使用旨在重建植被状况序列上包络线的平滑处理,我们获得了一个日植被数据集,该数据集有效地减轻了云层引起的波动,在捕捉区域气候方面优于极轨(如 MODIS)卫星数据集。我们通过统计分析证明了这一点,包括自相关性和连续观测之间的平均绝对差值。我们的研究表明,云污染会严重影响高时间分辨率的 NDVI 序列,尤其是在森林地区,这使得我们很难找到合适的数据集来验证我们的方法。因此,我们使用最大值合成技术来缓解这一问题,该技术旨在消除云层引起的偏差,并将我们的结果与从高级甚高分辨率辐射计获得的另一个时间分辨率更高的独立植被指数进行了进一步比较。我们发现,在剔除受云层污染的像素后,我们的植被指数与 MODIS 10 天合成图密切相关。此外,研究还评估了所选 NDVI 数据集对干旱事件的敏感性,证明了所建议的 SEVIRI 数据集在捕捉植被异常的强度和持续性方面的优势。总之,该研究提出了一种在云雾多发地区获取日分辨率 NDVI 数据集的创新策略,并利用不同次月时间尺度的独立数据集对其进行了验证。
{"title":"A high temporal resolution NDVI time series to monitor drought events in the Horn of Africa","authors":"Riccardo D’Ercole ,&nbsp;Daniele Casella ,&nbsp;Giulia Panegrossi ,&nbsp;Paolo Sanò","doi":"10.1016/j.jag.2024.104264","DOIUrl":"10.1016/j.jag.2024.104264","url":null,"abstract":"<div><div>This study investigates the reconstruction of climatological patterns and vegetation dynamics in the Horn of Africa region using high temporal resolution (i.e. daily) Normalized Difference Vegetation Index (NDVI) datasets. The analysis compares a straight-forward processing approach to derive a daily vegetation index from a geostationary (SEVIRI) satellite with existing NDVI series from geostationary or polar-orbiting (MODIS, MetOp-AVHRR) satellites, highlighting the impact of cloud contamination on data quality in high temporal resolution datasets. Using a smoothing process designed to reconstruct the upper envelope of the vegetation status series, we obtained a daily vegetation dataset that effectively mitigated cloud-induced fluctuations, outperforming polar-orbiting (e.g. MODIS) satellite-derived dataset in capturing regional climatology. We demonstrated this through statistical analysis, including autocorrelation and mean absolute difference between consecutive observations. We showed that cloud contamination significantly affects high temporal resolution NDVI series, particularly in forest areas, which makes it difficult to identify a suitable dataset to validate our approach. Therefore, we mitigated this problem using a Maximum Value Compositing technique, designed to remove cloud-induced biases and further compared our results with another independent vegetation index at coarser temporal resolution derived from AVHRR. We found that our vegetation index closely relates with MODIS 10-day composites after removing cloud-contaminated pixels. Furthermore, the study evaluates the sensitivity of the selected NDVI datasets to drought events, demonstrating the strength of the proposed SEVIRI dataset in capturing the intensity and persistence of vegetation anomalies. In conclusion, the study presents an innovative strategy for deriving daily-resolution NDVI datasets in cloud-prone regions, validating it with independent datasets at different sub-monthly temporal scales.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"135 ","pages":"Article 104264"},"PeriodicalIF":7.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142651954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
International journal of applied earth observation and geoinformation : ITC journal
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1