共检索到2047条,权限内显示50条;
[学术文献 ] Crop type mapping with temporal sample migration 进入全文
International Journal of Remote Sensing
Accurate and timely crop maps are crucial for monitoring agricultural production. Current supervised classification methods based on remote sensing rely heavily on ground-truth samples collected at a high cost, and the years without sampling highly limit classification accuracy. To address such a challenge, we proposed a time-migration method based on historical training samples collected in 2017, 2018 and 2020 to conduct supervised crop classification mapping in the target year (2021) with no ground samples. We chose Hailun City, Heilongjiang Province of northeastern China, as the study site; the major crops included corn, soybean, and rice. We reconstructed time series of Sentinel-2 data and selected spectro-temporal features to identify standard crop phenological curves. We calculated the similarity between reference and image spectra and designed label-matching rules to identify training samples through the dynamic time warping algorithm. We then used the historical samples to map the crop types of the target year. The results showed that the migration accuracy reached 95% for major crop. Using these samples as training data with a random forest to classify the target year, the overall accuracy reached 94.13%. The new sample time-migration method proposed in this study can efficiently migrate historical samples, greatly reducing the cost of ground-truth sampling.
[学术文献 ] Estimating fractional vegetation cover and aboveground biomass for land degradation assessment in eastern Mongolia steppe: combining ground vegetation data and remote sensing 进入全文
International Journal of Remote Sensing
Fractional vegetation cover (FVC) and aboveground biomass (AGB) are critically important for monitoring grassland degradation, and their accurate estimation can be used as key proxies for assessing land degradation. The main purpose of this study was to estimate the FVC and AGB in the eastern Mongolian steppe using remote sensing and machine learning. In this context, spectral bands and vegetation indices were extracted from the processed Sentinel-2 data and used as predictors. The field vegetation data were derived from the Mongolian pasture-monitoring database, which consisted of 256 plots with FVC and AGB measurements. Consequently, we derived FVC and AGB from Sentinel-2 imagery using 256 field vegetation measurements in the vast eastern Mongolian steppe as a reference for random forest (RF) models (R2FVC = 0.81, R²AGB = 0.76). Among the variables, the predictor variables derived from spectral vegetation and soil indices, especially NDVI, Simple Ratio (SR), and OSAVI, were highly important for predicting FVC and AGB. As expected, a comparison among the map values showed that the spatial distribution of FVC and AGB was consistent with the landscapes and ecoregions in the study area. As the FVC and AGB maps only showed the current condition of vegetation cover, we also analysed NDVI trends to explain vegetation cover changes. We tested temporal trends in vegetation using Landsat NDVI time series data and the Mann-Kendall trend test. This revealed that in 7.3% of the area, the NDVI significantly increased, whereas a significant decrease was observed in 58% of the area.
[科技报告 ] 2023全球粮食政策报告：重新思考粮食危机应对措施 进入全文
IFPRI
2022年，世界面临多重危机。旷日持久的2019冠状病毒病疫情(COVID-19)、重大自然灾害、内乱和政治动荡以及气候变化日益严重的影响对食物系统的破坏仍在继续，而与此同时，俄乌战争和通货膨胀加剧了全球粮食和化肥危机。危机数量不断增加，多种危机的叠加影响日益加剧，饥饿人口和流离失所者数量不断攀升，促使人们呼吁重新思考粮食危机应对措施，从而为变革创造了一个真正的机会。
[前沿资讯 ] Towards smarter agriculture: automatic identification of crop heads with artificial intelligence 进入全文
EurekAlert
Recent advances in artificial intelligence (AI), alongside drones and digital cameras, have greatly extended the frontiers of smart agriculture. One attractive use case for these technologies is precision agriculture. In this modern approach to farming, the idea is to optimize crop production by gathering precise data about plants and the state of the field, and then act accordingly. For example, by analyzing aerial images of crops, AI models can determine what parts of a field need more attention, as well as the current stage of development of the plants. Among all the crop monitoring functions that AI can do, crop head counting remains as one of the most challenging to implement. Images of crops contain densely packed, repeating patterns that are usually irregular and overlapped, making it difficult for deep learning models to automatically detect specific plant organs. Ideally, one would train such models using thousands of manually annotated images, in which pixels belonging to crop heads are pre-specified. In practice, however, annotating crop images is extremely tedious and time-consuming.
[前沿资讯 ] New grant to reveal tillage effects on crop yield, farmland sustainability 进入全文
伊利诺伊大学
Researchers from the Agroecosystem Sustainability Center (ASC) at the University of Illinois can detect soil tillage practices from space, weaving together data from ground images, airborne sensors, and satellites. Now, with a grant from the USDA’s National Institute for Food and Agriculture, they will expand on that work to produce more accurate estimates of tillage effects on corn and soybean yield, greenhouse gas emissions, nitrogen loss, and changes in soil organic carbon. Leading the project is Bin Peng, senior research scientist at ASC and research assistant professor in the Department of Natural Resources and Environmental Sciences (NRES) in the College of Agricultural, Consumer and Environmental Sciences (ACES) at Illinois. He says although no-till and other conservation tillage practices are on the rise throughout the U.S. Midwest, small-scale studies on tillage effects have produced contradictory results. Furthermore, no integrated high-resolution study has been done at large spatial scales.
[学术文献 ] Research Data Management Commitment Drivers: An Analysis of Practices, Training, Policies, Infrastructure, and Motivation in Global Agricultural Science 进入全文
ICARDA
Scientists largely acknowledge the value of research data management (RDM) to enable reproducibility and reuse. But, RDM practices are not sufficiently rewarded within the traditional academic reputation economy. Recent work showed that emerging RDM tools can offer new incentives and rewards. But, the design of such platforms and scientists' commitment to RDM is contingent on additional factors, including policies, training, and several types of personal motivation. To date, studies focused on investigating single or few of those RDM components within a given environment. In contrast, we conducted three studies within a global agricultural science organization, to provide a more complete account of RDM commitment drivers: one survey study (n = 23) and two qualitative explorations of regulatory frameworks (n = 17), as well as motivation, infrastructure, and training components (n = 13). Based on the sum of findings, we contribute to the triangulation of a recent RDM commitment evolution model. In particular, we find that strong support and suitable tools help develop RDM commitment, while policy conflicts, unclear data standards, and multi-platform sharing, lead to unexpected negotiation processes. We expect that these findings will help to better understand RDM commitment drivers, refine the RDM commitment evolution model, and benefit its application in science.
