A Spatial Multi-Criteria Decision-Making Approach to Evaluating Homogeneous Areas for Rainfed Wheat Yield Assessment

Water, Journal Year: 2025, Volume and Issue: 17(7), P. 1045 - 1045

Published: April 2, 2025

Rainfed wheat plays a vital role in global food security, particularly regions where water availability is limiting factor. Identifying homogeneous areas with similar yield potential essential for optimizing resource allocation, improving agricultural sustainability, and enhancing management. Unlike previous studies that primarily focused on cropland suitability, this study presents an integrated approach to delineate the rainfed using advanced mechanistic analysis multi-criteria decision-making techniques. Additionally, it examines homogeneity of these terms actual relative yield. Kurdistan province Iran was selected as area. Key phenological stages growth—germination, flowering, seed filling—were determined day-growth model. A set four primary criteria—precipitation, temperature, soil properties, topography—along twenty sub-criteria were based expert knowledge research. The Fuzzy-AHP method employed assign weights each factor, weighted linear combination used generate final classification map. results categorized area into five suitability classes: currently unsuitable (N2 N1), somewhat suitable (S3), moderately (S2), very (S1), accordance FAO standard framework. These classifications highlighted significant variations among zones. findings revealed highest lowest average yields observed classes S1 N2, respectively, yield-to-potential ratios ranging from 75% 20% N2. This research underscores spatial precision agriculture management, contributing more resilient production systems water-scarce regions.

Language: Английский

---

The Change Pattern and Its Dominant Driving Factors of Wetlands in the Yellow River Delta Based on Sentinel-2 Images

Remote Sensing, Journal Year: 2022, Volume and Issue: 14(17), P. 4388 - 4388

Published: Sept. 3, 2022

There were significant differences in the dominant driving factors of change process different types wetlands Yellow River delta. In addition, to our knowledge, optimal classification feature sets with Random Forest algorithm for delta least explored. this paper, wetland information study area was extracted based on a de-feature variable redundancy, and then its from 2015 2021 monitored analyzed using Geodetector gravity center model. The results showed that (1) composed red edge indexes had highest accuracy, overall accuracy Kappa coefficient 95.75% 0.93. (2) During 2015–2021, large natural transformed into an artificial wetland. development direction “northwest–southeast” along River. (3) interaction between vegetation coverage accumulated temperature largest explanatory power area. solar radiation DEM research could better provide decisions protection restoration

Language: Английский

---

Innovative Decision Fusion for Accurate Crop/Vegetation Classification with Multiple Classifiers and Multisource Remote Sensing Data

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(9), P. 1579 - 1579

Published: April 29, 2024

Obtaining accurate and real-time spatial distribution information regarding crops is critical for enabling effective smart agricultural management. In this study, innovative decision fusion strategies, including Enhanced Overall Accuracy Index (E-OAI) voting the Index-based Majority Voting (OAI-MV), were introduced to optimize use of diverse remote sensing data various classifiers, thereby improving accuracy crop/vegetation identification. These strategies utilized integrate classification outcomes from distinct feature sets (including Gaofen-6 reflectance, Sentinel-2 time series vegetation indices, biophysical variables, Sentinel-1 backscatter coefficients, their combinations) using classifiers (Random Forests (RFs), Support Vector Machines (SVMs), Maximum Likelihood (ML), U-Net), taking two grain-producing areas (Site #1 Site #2) in Haixi Prefecture, Qinghai Province, China, as research area. The results indicate that employing U-Net on feature-combined yielded highest overall (OA) 81.23% 91.49% #2, respectively, single classifier experiments. E-OAI strategy, compared original OAI boosted OA by 0.17% 6.28%. Furthermore, OAI-MV strategy achieved 86.02% 95.67% respective study sites. This highlights strengths features discerning different crop types. Additionally, proposed effectively harness benefits multisource features, significantly enhancing classification.

Language: Английский

---

AGTML: A novel approach to land cover classification by integrating automatic generation of training samples and machine learning algorithms on Google Earth Engine

Ecological Indicators, Journal Year: 2023, Volume and Issue: 154, P. 110904 - 110904

Published: Sept. 1, 2023

The timely, accurate, and automatic acquisition of land cover (LC) information is a prerequisite for detecting LC dynamics performing ecological analyses. Cloud computing platforms, such as the Google Earth Engine, have substantially improved efficiency scale classification. However, lack sufficient representative training samples hinders accurate In this study, we propose new approach that integrates generation machine learning algorithms (AGTML) classification in Heilongjiang Province, China. After optimal focal radii were determined different types using Landsat 8 based on statistics unique phenology. Then target automatically generated distance measure SED (a composite Spectral angle (SAD) Euclidean (ED)). Furthermore, was performed four feature combinations three algorithms. According to independent validation data, demonstrated good representativeness stability among all classifiers, with an overall accuracy (OA) higher than 86%, showed high consistency landscape pattern RF yielded highest (92.99% OA). AGTML outperformed GLC-FCS30 identifying large fragmentation small patch regions types. subsequently applied Guanzhong Plain satellite imagery. Results consistent (>96.50% OA), demonstrating can be various sensors, has immense potential automated across regional global scales.

Language: Английский

---

Improving grassland classification accuracy using optimal spectral-phenological-topographic features in combination with machine learning algorithm

Ecological Indicators, Journal Year: 2023, Volume and Issue: 158, P. 111392 - 111392

Published: Dec. 12, 2023

Accurate mapping of large-scale grassland types is important for and water resources management. The similarity spectral characteristics between lowers the classification accuracy different types. To improve grasslands, this study proposed a new framework which integrates Sentinel-2 images with DEM climate zones data. In framework, optimal spectral-phenological-topographic features are fed into Random Forest (RF) model based on Google Earth Engine (GEE) platform. was applied in Inner Mongolia, China. A map region obtained an overall (OA) exceeding 80 %, higher than OA (60 %-70 %) current type studies. WIM (Western Mongolia) NEIM (Northeast Mongolia), reaches 96.97 % 95.85 respectively. SWIR2 band elevation have clear advantage distinguishing Compared to 1980s, area temperate meadow steppe (TMS) desert (TDS) increased by 111.94 126.00 typical (TTS) decreased 7.38 %. has great potential be other regions future.

Language: Английский

---

Research on temporal and spatial evolution of land use and landscape pattern in Anshan City based on GEE

Frontiers in Environmental Science, Journal Year: 2022, Volume and Issue: 10

Published: Aug. 30, 2022

Frequent mining activities can bring about problems such as soil erosion and environmental pollution, which are detrimental to the efficient use of land sustainable development cities. Existing studies have paid little attention areas lack comparative analysis landscape changes in multiple pits. In this paper, main urban area Anshan City, where concentrated, was used research area, Landsat TM/OLI surface reflectance (SR) data Google Earth Engine (GEE) platform random forest algorithm were map 2008, 2014, 2020. On basis, dynamics pattern indices analyze patterns City area. addition, a moving window method combined further compare between different The results show that:1. From 2008 2020, construction continued decline, expand, shifted cultivated land. Mining increased before 2014 remained almost unchanged after is line with actual situation. 2. During study period, fragmentation degree heterogeneity kept increasing. high value urban-rural combination areas. Among them, reclamation Dagushan Donganshan better, while Anqian, Yanqianshan Xiaolingzi mines needs be strengthened. 3. based on GEE shows accuracy for classification. overall classification 3 years exceeds 90% kappa coefficient 0.85. an essential reference optimizing ecological environment provide technical backing urbanization rational City.

Language: Английский

---

Land use change detection and quantification of prime agricultural lands in Southern Ontario

Geoderma Regional, Journal Year: 2024, Volume and Issue: 37, P. e00775 - e00775

Published: Feb. 2, 2024

Spatiotemporal land use change evaluation and quantification are essential for supporting enhancing policy managing resources sustainably. To quantify the semi-decadal loss of prime agricultural lands in Southern Ontario, firstly, after pre-processing Agriculture Agri-Food Canada (AAFC) data, maps matrices spatial temporal converted to built-up were prepared. Then, changes capable soils lost due conversion was evaluated. Finally, amount soil different regions periods calculated. The area has increased decreased by 68% 4% from 1990 2020, respectively. period 1990–2005 2005–2020 1179 1640 km2, In these periods, 920 1204 km2 highly agriculture have been physical growth lands, results showed that rate areas over recent years.

Language: Английский

---

In-Depth Analysis and Characterization of a Hazelnut Agro-Industrial Context through the Integration of Multi-Source Satellite Data: A Case Study in the Province of Viterbo, Italy

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(7), P. 1227 - 1227

Published: March 30, 2024

The production of “Nocciola Romana” hazelnuts in the province Viterbo, Italy, has evolved into a highly efficient and profitable agro-industrial system. Our approach is based on hierarchical framework utilizing aggregated data from multiple temporal sources, offering valuable insights spatial, temporal, phenological distributions hazelnut crops To achieve our goal, we harnessed power Google Earth Engine utilized collections satellite images Sentinel-2 Sentinel-1. By creating dense stack multi-temporal images, precisely mapped groves area. During testing phase model pipeline, achieved an F1-score 99% by employing Hierarchical Random Forest algorithm conducting intensive sampling using high-resolution imagery. Additionally, employed clustering process to further characterize identified areas. Through this process, unveiled distinct regions exhibiting diverse spectral, responses. We successfully delineated actual extent cultivation, totaling 22,780 hectares, close accordance with national statistics, which reported 23,900 hectares total 21,700 for year 2022. In particular, three geographic distribution patterns orchards confined within PDO (Protected Designation Origin)-designated region. methodology pursued, years aggregate one SAR spectral separation approach, effectively allowed identification specific perennial crop, enabling deeper characterization various aspects influenced environmental configurations agronomic practices.The accurate mapping open opportunities implementing precision agriculture strategies, thereby promoting sustainability maximizing yields thriving

Language: Английский

---

SLMFNet: Enhancing land cover classification of remote sensing images through selective attentions and multi-level feature fusion

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0301134 - e0301134

Published: May 14, 2024

Land cover classification (LCC) is of paramount importance for assessing environmental changes in remote sensing images (RSIs) as it involves assigning categorical labels to ground objects. The growing availability multi-source RSIs presents an opportunity intelligent LCC through semantic segmentation, offering a comprehensive understanding Nonetheless, the heterogeneous appearances terrains and objects contribute significant intra-class variance inter-class similarity at various scales, adding complexity this task. In response, we introduce SLMFNet, innovative encoder-decoder segmentation network that adeptly addresses challenge. To mitigate sparse imbalanced distribution RSIs, incorporate selective attention modules (SAMs) aimed enhancing distinguishability learned representations by integrating contextual affinities within spatial channel domains compact number matrix operations. Precisely, position module (SPAM) employs pyramid pooling (SPP) resample feature anchors compute affinities. tandem, (SCAM) concentrates on capturing channel-wise affinity. Initially, maps are aggregated into fewer channels, followed generation pairwise between channels all channels. harness fine-grained details across multiple multi-level fusion decoder with data-dependent upsampling (MLFD) meticulously recover merge diverse scales using trainable projection matrix. Empirical results ISPRS Potsdam DeepGlobe datasets underscore superior performance SLMFNet compared state-of-the-art methods. Ablation studies affirm efficacy precision SAMs proposed model.

Language: Английский

---

Comparative Analysis of Multispectral and Hyperspectral Imagery for Mapping Sugarcane Varieties

PFG – Journal of Photogrammetry Remote Sensing and Geoinformation Science, Journal Year: 2023, Volume and Issue: 91(6), P. 453 - 470

Published: Sept. 6, 2023

Language: Английский

---